U.S. patent number 8,232,228 [Application Number 10/538,274] was granted by the patent office on 2012-07-31 for method for increasing the efficacy of agricultural chemicals.
This patent grant is currently assigned to Plant Health Care, Inc.. Invention is credited to Zhong-Min Wei.
United States Patent |
8,232,228 |
Wei |
July 31, 2012 |
Method for increasing the efficacy of agricultural chemicals
Abstract
The present invention is directed to increasing the efficacy of
agricultural chemicals. This can be achieved by applying at least
one agricultural chemical to a plant or plant seed under conditions
effective for the agricultural chemical to perform its intended
function and applying at least one hypersensitive response elicitor
protein or polypeptide to the plant or plant seed under conditions
effective to increase the efficacy of the agricultural chemical.
Alternatively, the present invention relates to a method for
increasing the efficacy of agricultural chemicals by applying an
agricultural chemical to a transgenic plants or transgenic seeds
transformed with nucleic acid molecule which encodes a
hypersensitive response elicitor protein or polypeptide, wherein
the agricultural chemical is applied under conditions effective for
the agricultural chemical to perform its intended function but with
increased efficacy.
Inventors: |
Wei; Zhong-Min (Kirkland,
WA) |
Assignee: |
Plant Health Care, Inc.
(Pittsburgh, PA)
|
Family
ID: |
32681976 |
Appl.
No.: |
10/538,274 |
Filed: |
December 15, 2003 |
PCT
Filed: |
December 15, 2003 |
PCT No.: |
PCT/US03/40089 |
371(c)(1),(2),(4) Date: |
September 28, 2006 |
PCT
Pub. No.: |
WO2004/057957 |
PCT
Pub. Date: |
July 15, 2004 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20070037705 A1 |
Feb 15, 2007 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
60433893 |
Dec 16, 2002 |
|
|
|
|
Current U.S.
Class: |
504/100; 504/118;
504/116.1 |
Current CPC
Class: |
A01N
63/50 (20200101); A01N 47/34 (20130101); A01N
57/20 (20130101); A01N 47/34 (20130101); A01N
43/76 (20130101); A01N 47/34 (20130101); A01N
2300/00 (20130101); A01N 57/20 (20130101); A01N
2300/00 (20130101); A01N 63/50 (20200101); A01N
63/20 (20200101); A01N 63/50 (20200101); A01N
57/28 (20130101); A01N 57/12 (20130101); A01N
57/20 (20130101); A01N 47/24 (20130101); A01N
47/34 (20130101); A01N 43/76 (20130101); A01N
61/00 (20130101); A01N 63/50 (20200101); A01N
63/27 (20200101); A01N 63/50 (20200101); A01N
43/76 (20130101); A01N 47/24 (20130101); A01N
47/34 (20130101); A01N 57/12 (20130101); A01N
57/20 (20130101); A01N 57/28 (20130101); A01N
59/06 (20130101); A01N 57/20 (20130101); A01N
63/50 (20200101); A01N 63/50 (20200101); A01N
2300/00 (20130101) |
Current International
Class: |
A01N
25/26 (20060101); A01N 25/00 (20060101); A01N
63/00 (20060101) |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
WO 96/23411 |
|
Aug 1996 |
|
WO |
|
WO 98/37752 |
|
Sep 1998 |
|
WO |
|
WO-9837752 |
|
Sep 1998 |
|
WO |
|
WO 99/35913 |
|
Jul 1999 |
|
WO |
|
WO 01/67865 |
|
Sep 2001 |
|
WO |
|
Other References
Bailey et al., "Factors Influencing the Herbicidal Activity of
Nep1, a Fungal Protein That Induces the Hypersensitive Response in
Centaurea maculosa," Weed Science 48(6):776-785 (2000). cited by
other .
Tosun et al., "The Effect of Harpin EA as Plant Activator in
Control of Bacterial and Fungal Diseases of Tomato," Acta Hort.
613:251-254 (2003). cited by other .
Supplementary European Search Report dated Jan. 5, 2010. cited by
other.
|
Primary Examiner: Pryor; Alton
Attorney, Agent or Firm: LeClairRyan, a Professional
Corporation
Parent Case Text
This application is a national stage application under 35 U.S.C.
.sctn.371 from PCT Application No. PCT/US03/40089, filed Dec. 15,
2003, which claims the priority benefit of U.S. Provisional Patent
Application No. 60/433,893, filed Dec. 16, 2002.
Claims
What is claimed is:
1. A method of treating at least one plant or plant seed with at
least one insecticide, fungicide, or herbicide where a local
population of pests has resistance to the at least one insecticide,
fungicide, or herbicide, said method comprising: selecting the at
least one plant or plant seed to be treated by the at least one
insecticide, fungicide, or herbicide under conditions effective for
the at least one insecticide, fungicide, or herbicide to perform
its intended function, wherein the selected at least one plant or
plant seed is planted where the local population of pests has
resistance to the at least one insecticide, fungicide, or herbicide
and applying the at least one insecticide, fungicide, or herbicide
and at least one hypersensitive response elicitor protein or
polypeptide to said selected at least one plant or plant seed under
conditions effective to treat the at least one plant or plant seed
with the at least one insecticide, fungicide, or herbicide, wherein
said at least one hypersensitive response elicitor is heat stable,
glycine rich, and contains substantially no cysteine.
2. The method according to claim 1, wherein said selected plant is
treated during said applying.
3. The method according to claim 1, wherein said selected plant
seed is treated during said applying, said method further
comprising: planting said treated, selected plant seed in natural
or artificial soil and propagating a plant from said treated,
selected plant seed planted in said natural or artificial soil.
4. The method according to claim 1, wherein said selected plants or
plant seeds are selected from the group consisting of canola,
alfalfa, rice, wheat, barley, rye, cotton, sunflower, peanut, corn,
potato, sweet potato, bean, pea, chicory, lettuce, endive, cabbage,
brussel sprout, beet, parsnip, cauliflower, broccoli, turnip,
radish, spinach, onion, garlic, eggplant, pepper, celery, carrot,
squash, pumpkin, zucchini, cucumber, apple, pear, melon, citrus,
strawberry, grape, raspberry, pineapple, soybean, tobacco, tomato,
sorghum, avocado, sugarcane, Saintpaulia, petunia, pelargonium,
poinsettia, chrysanthemum, carnation, and zinnia.
5. The method according to claim 1, wherein said applying the at
least one insecticide, fungicide, or herbicide is conducted
simultaneously or independently of said applying the at least one
hypersensitive response elicitor protein or polypeptide.
6. The method according to claim 1, wherein the at least one
insecticide is applied, said insecticide containing an active
ingredient selected from the group consisting of carbamates,
organochlorines, nicotinoids, phosphoramidothioates,
organophosphates, and pyrethroids.
7. The method according to claim 1, wherein the at least one
fungicide is applied, said fungicide containing an active
ingredient selected from the group consisting of aliphatic
nitrogens, benzimidazoles, dicarboximides, dithiocarbamates,
imidazoles, strobins, anilides, aromatics, sulfur derivatives, and
copper derivatives.
8. The method according to claim 1, wherein the at least one
herbicide is applied, said herbicide is selected from the group
consisting of acetyl-CoA carboxylase inhibitors (ACCase),
actolactate synthase inhibitors (ALS), microtubule assembly
inhibitors (MT), growth regulators (GR), photosynthesis II, binding
site A inhibitors (PSII(A)), photosynthesis II, binding site B
inhibitors (PSII(B)), photosynthesis II, binding site C inhibitors
(PSII(C)), shoot inhibitors (SHT), enolpyruvyl-shikimate-phosphate
synthase inhibitors (EPSP), glutamine synthase inhibitors (GS),
phytoene desaturase synthase inhibitors (PDS), diterpene inhibitors
(DITERP), protoporphyrinogen oxidase inhibitors (PPO), shoot and
root inhibitors (SHT/RT), photosystem 1 electron diverters (ED),
hydroxyphenlypyruvate dioxygenase synthesis inhibitors (HPPD), and
combinations thereof.
9. The method according to claim 1, wherein the at least one
hypersensitive response elicitor or polypeptide is derived from a
species of pathogens selected from the group consisting of Erwinia,
Pseudomonas, and Xanthomonas.
10. The method according to claim 9, wherein the at least one
hypersensitive response elicitor protein or polypeptide is derived
from an Erwinia species selected from the group consisting of
Erwinia amylovora, Erwinia carotovora, Erwinia chrysanthemi, and
Erwinia stewartii.
11. The method according to claim 9, wherein the at least one
hypersensitive response elicitor protein or polypeptide is derived
from a Pseudomonas species selected from the group consisting of
Pseudomonas syringae and Pseudomonas solanacearum.
12. The method according to claim 9, wherein the at least one
hypersensitive response elicitor or polypeptide is derived from
Xanthomonas campestris.
13. The method according to claim 1, wherein the at least one
insecticide is applied, said insecticide comprising nicotinoid.
14. The method according to claim 1, wherein the at least one
fungicide is applied, said fungicide comprising strobin.
15. The method according to claim 1, wherein the at least one
herbicide is applied, said herbicide comprising glyphosate.
16. The method according to claim 1, wherein the at least one
herbicide and the at least one fungicide are applied, said
herbicide comprising glyphosate and said fungicide comprising
strobin.
17. The method according to claim 1, wherein the at least one
herbicide and the at least one insecticide are applied, said
herbicide comprising glyphosate and said insecticide comprising
nicotinoid.
18. The method according to claim 1, wherein the at least one
herbicide is applied, said herbicide comprising glyphosate and
Dicamba.
19. The method according to claim 1, wherein the at least one
herbicide and the at least one fungicide are applied, said
herbicide comprising glyphosate, and Dicamba, and said fungicide
comprising strobin.
20. The method according to claim 1, wherein the at least one
herbicide and the at least one insecticide are applied, said
herbicide comprising glyphosate, and Dicamba, and said insecticide
comprising nicotinoid.
21. The method according to claim 8, wherein the at least one
herbicide is applied, said herbicide comprising a
enolpyruvyl-shikimate-phosphate synthase inhibitor (EPSP)
glyphosate.
22. The method according to claim 6, wherein the at least one
insecticide is applied, said insecticide comprising a
pyrethroid.
23. The method according to claim 7, wherein the at least one
fungicide is applied, said fungicide comprising a
benzimidazole.
24. A method of treating at least one transgenic plant or
transgenic seed with at least one insecticide, fungicide, or
herbicide where a local population of pests has resistance to the
at least one insecticide, fungicide, or herbicide, said method
comprising: selecting the at least one transgenic plant or
transgenic seed, transformed with at least one nucleic acid
molecule which encodes at least one hypersensitive response
elicitor protein or polypeptide, to be treated by the at least one
insecticide, fungicide, or herbicide under conditions effective for
the at least one insecticide, fungicide, or herbicide to treat at
least one pest, wherein the selected at least one transgenic plant
or transgenic seed is planted where the local population of pests
has resistance to the at least one insecticide, fungicide, or
herbicide and applying the at least one insecticide, fungicide, or
herbicide to said selected transgenic plant or transgenic seed
under conditions effective to treat the at least one transgenic
plant or transgenic seed with the at least one insecticide,
fungicide, or herbicide and for the at least one insecticide,
fungicide, or herbicide to perform its intended functions, wherein
said at least one hypersensitive response elicitor is heat stable,
glycine rich, and contains substantially no cysteine.
Description
FIELD OF INVENTION
The present invention relates to methods of increasing the efficacy
of commonly used agricultural chemicals.
BACKGROUND
Modern agricultural practices rely heavily on the use of chemical
inputs to maintain and increase productivity. Agricultural chemical
inputs can be broadly categorized as pesticides, fertilizers, and
plant growth regulators. Based on monetary expenditure, as well as
physical volumes, the vast majority of chemical inputs are in the
form of pesticides and fertilizers. In the common agricultural
sense, pests are any organisms that contribute to a loss of value
or productivity in a crop. Pesticides can be categorized into;
insecticides, fungicides, herbicides, as well as minor categories
such as acaricides, avicides, virucides, and nematicides. In 1996,
U.S. farmers spent over $8.5 billion on pesticides. This translates
to the use of over 355 million pounds of herbicides, 70 million
pounds of insecticides, and 180 million pounds of fungicides and
other pesticides in 1996 alone (Fernandez-Conejo and Jans, "Pest
Management in the U.S. Agriculture." Resource Economics Division,
Economic Research Service, U.S. Department of agriculture.
Agricultural Handbook No. 717.). With some exceptions, fertilizers
are typically characterized as substances containing plant
macronutrients or plant micronutrients, and are used in as
proportionally as large of volumes as pesticides. In 1997,
approximately 22 million tons of nutrients were applied in the
United States alone (Data from the Economic Research Service, U.S.
Department of Agriculture). Plant growth regulators are a class of
agricultural chemical inputs whose use is minor compared to
pesticides and fertilizers. Nonetheless, plant growth regulators
have significant importance in specific agricultural sectors such
as fruit production and ornamentals.
Though the increase in use of agricultural chemicals has directly
contributed to an increase in productivity, the increased
productivity has not come without a price. Most pesticides present
inherent human and environmental health risks. Increasingly,
municipalities are identifying hazardous agricultural chemicals, or
their residues, in local water sources, streams, and lakes. In
addition, the high volumes of pesticides being applied results in
the development of pest resistance to the agricultural chemical
being applied. Incidences of pest resistance have been documented
in most classes of pesticide and a wide range of crop types.
Resistance occurs after persistent use of a pesticide or closely
related pesticides has decimated a local population of pests, but
left a small sub-population of the same pest surviving. The
sub-population, either through human pressure or natural divergence
of ecotypes, has evolved to be less affected or resistant to the
pesticide or closely related pesticides. After repeated cycles of
heavy use of the pesticide, decimation of the local population, and
survival of the resistant sub-populations, the resistant
sub-population eventually multiplies to become the dominant
population. The end result being, an entire pest population that is
resistant to a given pesticide or closely related pesticides. A
once effective and important pesticide is essentially rendered
useless to the farmer or commercial grower. Prior to recognition of
the actual existence of a resistant pest, the grower having
recognized a decrease in efficacy of a pesticide will often
intuitively increase the amount of pesticide being applied. Thus,
compounding the situation by furthering the propagation of
resistant pest through increased use of the pesticide, decreasing
the profitability of the crop because of increased purchases of
chemical inputs, and simultaneously increasing the human and
environmental health risks.
Greater crop yields, resulting from an increased use of
fertilizers, have not come without detrimental effects either.
Fertilizers are applied to cropland to replenish or add nutrients
that are needed by an existing or future crop. The vast majority of
the nutrients applied are in the form of nitrogen, phosphorus, and
potash (i.e. potassium). Depending on a combination of factor such
as the soil's chemical structure, pH, and texture; fertilizer
components can be highly susceptible to leaching. Leaching occurs
when the amount of water present in the soil, either from
irrigation of rainfall, is greater than the soil's water-holding
capacity. When this occurs, solubilized fertilizer components are
carried low into the soil and out of the plant root zone, thus
effectively removing the nutrients for use by the plant.
Nitrate-nitrogen (NO.sub.3.sup.-) is particularly prone to
leaching, and can result in hazardous nitrate accumulation in
groundwater. In the U.S. and abroad, cropland is commonly
over-fertilized. Soil nutrient analysis is often viewed as timely
and not economically feasible. Thus, fertilizers are often applied
at regular intervals regardless of their need. As with pesticides,
the over use of the fertilizers has potentially far reaching
detrimental effect on agricultural profitability and risk to
environmental health.
In recent years, farmers and agricultural researchers have begun to
develop programs and techniques to aid in combating the cycles of
increased chemical inputs and decreased profitability. These
programs and techniques are commonly referred to as Integrated Pest
Management (IPM), or more broadly, Integrated Crop Management
(ICM). ICM programs and techniques are being advanced by a range of
organizations including; the USDA, land-grant universities and the
private sector. ICM Programs are specifically designed with respect
to crop type, local environmental conditions, and local pest
pressures. In contrast to previous agricultural practices, ICM
practices draw on a broad range of techniques and tools including;
early and persistent monitoring of pest populations, establishment
of acceptable pest population thresholds, the development of
chemical control programs that routinely rotate the chemicals being
utilized, establishment of cultural control techniques (e.g.
adjusting planting and harvesting dates, no-till systems, crop
rotation, etc.), promotion of the use of specific plant varieties
or transgenic plants, and the development of biological controls
techniques (e.g. use of beneficial insects, use of pheromones
traps, use of live micro-organisms such as Bacillius thuringensis,
etc.). Although ICM practices show great promise for combating many
of the problems associated with the high chemical input of modern
agricultural practices, the ability to increase the efficacy of the
commonly used agricultural chemicals would greatly aid in the over
all effort. Increased efficacy would provide greater pest control
and/or facilitate decreases in the volume of agricultural chemicals
used.
As evident from the above discussion, modem agricultural practices
dictate the need to apply several agricultural chemicals, often
repeatedly, to a single crop over the course of a growing season.
To facilitate this need to apply numerous chemicals to a single
crop, it has become routine practice to make what is referred to as
tank mixes. Tank mixes are a single application of one or more
chemical at the same time. The agricultural chemicals that are
desired to be applied are combined into one tank, mixed,
soluablized if needed, and applied to the crop. There are
limitations to this practice in that some agricultural chemicals
are not compatible and may precipitate, become inactive, or
decrease the efficacy of other chemicals when mixed together.
Pesticide interactions are typically characterized as additive,
synergistic, antagonistic, and enhancement. Additive effects occur
when the combination of two pesticides produces the same amount of
control as the combined effects of each of the chemicals applies
independently. Synergistic effects occur when the combined effects
of the chemicals is greater than the additive effects. It is
assumed that in synergistic pesticide interactions the chemicals
are not neutral to one another, and to some extent are chemically
interacting with one another. Antagonistic effects are those
resulting when the combination of chemicals is less than if the
chemicals were used individually. Enhanced effects can occur when a
pesticide is combined with an additive that is not a pesticide and
the resulting control of the desired pest is greater than if the
pesticide was used individually. Factor such as the quantity of
water used, the order of mixing the chemicals, and the addition of
ajuvants may also affect the utility of a tank mix (Petroff,
"Pesticide Interaction and Compatibility," Montana State
University).
The present invention is directed towards improving the efficacy of
agricultural chemicals.
SUMMARY OF THE INVENTION
The present invention relates to a method for increasing the
efficacy of agricultural chemicals. In one embodiment of the
present invention, the method is carried out by applying at least
one agricultural chemical and at least one least one hypersensitive
response elicitor protein or peptide to a plant or plant seed under
conditions effective to increase the efficacy of the agricultural
chemical.
In addition, the present invention relates to a method for
increasing the efficacy of agricultural chemicals by applying one
or more agricultural chemicals to a transgenic plants or transgenic
seeds transformed with a nucleic acid molecule which encodes a
hypersensitive response elicitor protein or polypeptide under
conditions effective for the agricultural chemical to perform its
intended function but with increased efficacy.
By the present invention, the efficacy of an agricultural chemical
is increased. In achieving this objective, the present invention
enables the grower to more effectively manage their crops with
respect to fertilizers and plant growth regulators and to more
effectively control pests such as insects, fungus, disease, and
weeds. As a result of the increased efficacy in controlling common
pest problems, growers can reduce yield losses resulting from pest
problems. In addition, the present invention enables growers to
utilize lower quantities of commonly utilized agricultural
chemicals while maintaining or increasing yields. The reduction of
agricultural chemical use will also result in profound health and
ecological benefits.
DETAILED DESCRIPTION OF THE INVENTION
The present invention relates to a method for increasing the
efficacy of agricultural chemicals. In one embodiment of the
present invention, the method is carried out by applying at least
one agricultural chemical and at least one least one hypersensitive
response elicitor protein or peptide to a plant or plant seed under
conditions effective to increase the efficacy of the agricultural
chemical.
Agricultural chemicals, according to the present invention, can be
divided into several broad categories pesticides, fertilizers, and
plant growth regulators. Pesticides, perhaps the most diverse
category of agricultural chemicals, can be subdivided into
categories based on the type of pest or organism which they are
intended to control, such as; insecticides, intended for the
control of insect; fungicides, intended for the control of fungi;
herbicides, intended for the control of noxious weeds and plants;
acaricides, intended for the control of arachnids or spiders;
virucides intended for the control of viruses; and nematicides,
intended for the control of nematodes.
For use in accordance with this method, suitable insecticides
include but, are not limited to those listed in Table 1.
TABLE-US-00001 TABLE 1 Common Agricultural Insecticides Common Name
of Class of Active Active Example Ingredient Ingredient Active
ingredient Product Name carbamate Aldricarb
2-methyl-2-(methylthio)propanal O- Temik .RTM. (Aventis (ISO)
[(methylamino)carbonyl]oxime (CAS) CropScience, Research Triangle
Park, NC) organochlorine Endosulfan
6,7,8,9,10,10-hexachloro-1,5,5a,6,9,9a- Thiodan .RTM. (Aventis
(ISO) hexahydro-6,9-methano-2,4,3- CropScience, Research
benzodioxathiepin 3-oxide (CAS) Triangle Park, NC) nicotinoid
Imidacloprid 1-[(6-chloro-3-pyridinyl)methyl]-N- Merit .RTM. (Bayer
(ISO) nitro-2-imidazolidinimine (CAS) Ag, Leverkusen, Germany)
phosphoramidothioate Acephate O,S-dimethyl
acetylphosphoramidothioate Orthene .RTM. (Valent (ISO) (CAS) U.S.A.
Corp., Walnut Creek, CA) organothiophosphate Dimethoate
O,O-dimethyl S-[2-(methylamino)-2- Roxion .RTM. (BASF (ISO)
oxoethyl] phosphorodithioate (CAS) Corp., Research Triangle Park,
NC) pyrethroid Permethrin (3-phenoxyphenyl)methyl 3-(2,2- Ambush
.RTM. (ISO) dichloroethenyl)-2,2- (Syngenta, Greensboro NC)
dimethylcyclopropanecarboxylate (CAS)
Table 1 is intended as an example. Alternative example product
names and classifications exist for the active ingredients cited
and would fall within the scope of the present invention.
For use in accordance with this method, suitable fungicides include
those listed in Table 2. In addition to Table 2, suitable
fungicides include various forms of organic and inorganic copper.
Examples of suitable copper compounds include, copper ammonium,
copper hydroxide, copper oxychloride, and copper-zinc-chromate.
TABLE-US-00002 TABLE 2 Common Agricultural Fungicides Common Name
Class of Active of Active Example Product Ingredient Ingredient
Active ingredient Name aromatic Chlorothalonil
Tetrachloroisophthalonitrile (IUPAC) Bravo .RTM. (Syngenta, (ISO)
Greensboro NC) copper copper hydroxide copper hydroxide
(Cu(OH).sub.2) (CAS) Kocide .RTM. (Griffin L.L.C., Valdosta GA)
sulfur Flowers of Sulfur sulfur Kumulus .RTM. (BASF Corp., Research
Triangle Park, NC) aliphatic nitrogen Cymoxanil (ISO)
2-cyano-N-[(ethylamino)carbonyl]-2- Curzate .RTM. (DuPont
(methoxyimino)acetamide (CAS) Crop Protection, Wilmington, DE)
benzimidazole Thiabendazole 2-(4-thiazolyl)-1H-benzimidazole
Thiabendazole .RTM. (ISO) (CAS) (Syngenta, Greensboro NC)
dicarboximide Captan (ISO) 3a,4,7,7a-tetrahydro-2- Captan .RTM.
(Syngenta, [(trichloromethyl)thio]-1H-isoindole- Greensboro NC)
1,3(2H)-dione (CAS) dicarboximide Vinclozolin (ISO)
3-(3,5-dichlorophenyl)-5-ethenyl-5- Ronilan .RTM. (BASF
methyl-2,4-oxazolidinedione (CAS) Corp., Research Triangle Park,
NC) dithiocarbamate Mancozeb (ISO)
[[1,2-ethanediylbis[carbamodithioato]] Dithane .RTM. (Rohm and
(2-)]manganese mixture with [[1,2- Haas Co.,
ethanediylbis[carbamodithioato]](2-)] Philadelphia, PA) zinc (CAS)
dithiocarbamate Maneb (ISO) [[1,2- Manex .RTM. (Griffin
ethanediylbis[carbamodithioato]](2-)] L.L.C., Valdosta GA)
manganese (CAS) dithiocarbamate Metiram (JMAFF) zinc ammoniate
Polyram .RTM. (BASF ethylenebis(ditbiocarbamate) - Corp., Research
poly(ethylenethiuram disulfide) Triangle Park, NC) (IUPAC)
dithiocarbamate Thiram (ISO) tetramethylthioperoxydicarbonic Thiram
.RTM. (BASF diamide ([[(CH.sub.3).sub.2N]C(S)].sub.2S.sub.2) (CAS)
Corp., Research Triangle Park, NC) dithiocarbamate Ziram (ISO)
(T-4)-bis(dimethylcarbamoditbioato- Ziram .RTM. (UBC S,S')zinc
Agrochemicals, Ghent, Belgium) imidazole, Iprodione (ISO)
3-(3,5-dichlorophenyl)-N-(1- Rovral .RTM. (Aventis dicarboximide
methylethyl)-2,4-dioxo-1- CropScience, imidazolidinecarboxamide
(CAS) Research Triangle Park, NC) organophosphate Fosetyl-aluminum
ethyl hydrogen phosphonate(CAS) as Alientte .RTM. (Aventis (ISO) an
aluminum salt CropScience, Research Triangle Park, NC) strobin
Azoxystrobin (.alpha.E)-methyl 2-[[6-(2-cyanophenoxy)- Abound .RTM.
(Syngenta, (ISO) 4-pyrimidinyl]oxy]-.alpha.- Greensboro NC)
(methoxymethylene)benzeneacetate (CAS) anilide Metalaxyl (ISO)
methyl N-(2,6-dimethylphenyl)-N- Ridomil .RTM. (Syngenta,
(methoxyacetyl)-DL-alaninate (CAS) Greensboro NC)
Table 2 is intended as an example. Alternative example product
names and classifications exist for the active ingredients cited
and would fall within the scope of the present invention.
For use in accordance with this method, suitable herbicides
include, but are not limited to those listed in Tables 3 and 4.
Table 3 outlines a Site of Action Classification of Herbicides and
is based on the classification system developed by the Weed Science
Society of America (WSSA). The herbicide's site of action is
defined as the specific biochemical process in the plant that the
herbicide acts upon or disrupts. For example, an herbicide
containing the active ingredient primisulfuron, has a site of
action classification number 2. Table 3 indicates that a
Classification Number 2 has as its site of action acetolactate
synthase inhibition.
TABLE-US-00003 TABLE 3 Herbicide Site of Action and Classification
Numbers. Site of Action Classification No. Description of Site of
Action 1 ACCase = acetyl-CoA carboxylase inhibitor 2 ALS =
actolactate synthase inhibitor 3 MT = microtubule assembly
inhibitor 4 GR = growth regulator 5 PSII(A) = photosynthesis II,
binding site A inhibitor 6 PSII(B) = photosynthesis II, binding
site B inhibitor 7 PSII(C) = photosynthesis II, binding site C
inhibitor 8 SHT = shoot inhibitor 9 EPSP =
enolpyruvyl-shikimate-phosphate synthase inhibitor 10 GS =
glutamine synthase inhibitor 12 PDS = phytoene desaturase synthase
inhibitor 13 DITERP = diterpene inhibitor 14 PPO =
protoporphyrinogen oxidase inhibitor 15 SHT/RT = shoot and root
inhibitor 22 ED = photosystem 1 electron diverter 28 HPPD =
hydroxyphenlypyruvate dioxygenase synthesis inhibitor
TABLE-US-00004 TABLE 4 Common Agricultural Herbicides Common Name
of Site of Class of Active Active Example Action Ingredient
Ingredient Active ingredient Product Name 1 Cyclohexene Oxime
Sethoxydim 2-[1-(ethoxyimino)butyl]-5-[2- Poast .RTM. (BASF (ISO)
(ethylthio)propyl]-3-hydroxy-2- Corp., Research cyclohexen-1-one
(CAS) Triangle Park, NC) 1 Phenoxy Quizalofop-P
(R)-2-[4-[(6-chloro-2- Assure II .RTM. (ISO)
quinoxalinyl)oxy]phenoxy]propanoic (DuPont Crop acid (CAS)
Protection, Wilmington, DE) 2 Sulfonylurea Primisulfuron
2-[[[[[4,6-bis(difluoromethoxy)-2- Beacon .RTM. (ISO)
pyrimidinyl]amino]carbonyl]amino] (Syngenta, sulfonyl]benzoic acid
(CAS) Greensboro NC) 2 Imidazolinone Imazamox
2-[4,5-dihydro-4-methyl-4-(1- Raptor .RTM. (BASF (ISO)
methylethyl)-5-oxo-1H-imidazol-2- Corp., Research
yl]-5-(methoxymethyl)-3- Triangle Park, NC) pyridinecarboxylic acid
(CAS) 3 Dinitroaniline Trifluralin 2,6-dinitro-N,N-dipropyl-4-
Passport .RTM. (ISO) (trifluoromethyl)benzenamine (CAS) (BASF
Corp., Research Triangle Park, NC) 3 Dinitroaniline Pendimethalin
N-(1-ethylpropyl)-3,4-dimethyl-2,6- Prowl .RTM. (BASF (ISO)
dinitrobenzenamine (CAS) Corp., Research Triangle Park, NC) 4
Phenoxy 2,4-D (ISO) (2,4-dichlorophenoxy)acetic acid Amsol .RTM.
(CAS) (Aventis CropScience, Research Triangle Park, NC) 4 Benzoic
acid Dicamba 3,6-dichloro-2-methoxybenzoic acid Banvel .RTM. (BASF
(ISO) (CAS) Corp., Research Triangle Park, NC) 5 Triazine Atrazine
6-chloro-N-ethyl-N'-(1-methylethyl)- Atrazine .RTM. (ISO)
1,3,5-triazine-2,4-diamine (CAS) (Syngenta, Greensboro NC) 5
Triazine Cyanazine 2-[[4-chloro-6-(ethylamino)-1,3,5- Blandex .RTM.
(ISO) triazin-2-yl]amino]-2- (BASF Corp., methylpropanenitrile
(CAS) Research Triangle Park, NC) 6 Nitrite Bromoxylin
3,5-dibromo-4-hydroxybenzonitrile Buctril .RTM. (ISO) (CAS)
(Aventis CropScience, Research Triangle Park, NC) 7 Phenylurea
Diuron (ISO) N'-(3,4-dichlorophenyl)-N,N- Karmex .RTM. dimethylurea
(CAS) (Griffin L.L.C., Valdosta GA) 8 Thiocarbamate EPTC (ISO)
S-ethyl dipropylcarbamothioate (CAS) Eptam .RTM. (Syngenta,
Greensboro NC) 9 Organophosphorus Glyphosate
N-(phosphonomethyl)glycine (CAS) Roundup .RTM. (ISO) (Monsanto Co.,
St Louis MO) 10 Organophosphorus Glufosinate 2-amino-4- Liberty
.RTM. (ISO) (hydroxymethylphosphinyl)butanoic (Aventis acid (CAS)
CropScience, Research Triangle Park, NC) 12 Pyridazinone
Norflurazon 4-chloro-5-(methylamino)-2-[3- Zorial .RTM. (ISO)
(trifluoromethyl)phenyl]-3(2H)- (Syngenta, pyridazinone (CAS)
Greensboro NC) 13 unclassified Clomazone
2-[(2-chlorophenyl)methyl]-4,4- Command .RTM. (ISO)
dimethyl-3-isoxazolidinone (CAS) (FMC Corp., Philadelphia, PA) 14
Diphenyl ether Fomesafen 5-[2-chloro-4- Reflex .RTM. (ISO)
(trifluoromethyl)phenoxy]-N- (Syngenta,
(methylsulfonyl)-2-nitrobenzamide Greensboro NC) (CAS) 15
Chloroacetanilide Alachlor 2-chloro-N-(2,6-diethylphenyl)-N- Lasso
.RTM. (ISO) (methoxymethyl)acetamide (CAS) (Monsanto Co., St. Louis
MO) 15 Chloroacetanilide Acetochlor 2-chloro-N-(ethoxymethyl)-N-(2-
Surpass .RTM. (Dow (ISO) ethyl-6-methylphenyl)acetamide AgroScience
LLC, (CAS) Indianapolis, IN) 22 Quaternary Diquat (ISO)
6,7-dihydrodipyrido[1,2-.alpha.:2',1'- Reglone .RTM. ammonium
c]pyrazinediium (CAS) (Syngenta, Greensboro NC) 28
Cyclopropylisoxazole Isoxaflutole (5-cyclopropyl-4-isoxazolyl)[2-
Balan- ce .RTM. (ISO) (methylsulfonyl)-4- (Aventis
(trifluoromethyl)phenyl]methanone CropScience, (CAS) Research
Triangle Park, NC)
Table 4 is intended as an example. Alternative example product
names and classifications exist for the active ingredients cited
and would fall within the scope of the present invention.
For use in accordance with this method, suitable fertilizers
include, but are not limited to those containing plant
micronutrients (molybdenum, copper, zinc, manganese, iron, boron,
cobalt, and chlorine) and plant macronutrients (sulfur, phosphorus,
magnesium, calcium, potassium, and nitrogen). Numerous combinations
and forms of plant macro and micronutrients exist and are available
in a wide range of formulations. The predominant fertilizers used
in agriculture contain various combinations and concentrations of
nitrogen, phosphorus, and potassium. Micronutrient specific
fertilizers are also common and may contain a single micronutrient
or a combination of several micronutrients.
For use in accordance with this method, suitable plant growth
regulators include, but are not limited to those containing various
form and combinations of auxins, cytokinins, defoliants,
gibberellins, ethylene releaser, growth inhibitors, growth
retardants, and growth stimulators. Specific plant growth
regulators include but are not limited to those listed in Table
5.
TABLE-US-00005 TABLE 5 Common Plant Growth Regulators Class of
Active Common Name of Example Product Ingredient Active Ingredient
Active ingredient Name Cytokinin Zeatin
(E)-2-methyl-4-(1H-purin-6-ylamino)-2- buten-1-ol Defoliant
Thidiazuron (ISO) N-phenyl-N'-1,2,3-thiadiazol-5-ylurea Dropp .RTM.
(Aventis (CAS) CropScience, Research Triangle Park, NC) Growth
Forchlorfenuron N-(2-chloro-4-pyridinyl)-N'-phenylurea stimulator
(CAS) Growth Mepiquat (ISO) N,N-dimethylpiperdinum chloride (CAS)
Pix .RTM. (BASF Corp., Inhibitor chloride Research Triangle Park,
NC) Growth Maleic Hydrazide 1,2-dihydro-3,6-pyridazinedione (CAS)
Sprout Stop .RTM. Inhibitor (ISO-E) (Drexel Chemical Co., Memphis,
TN) Growth Palclobutrazol (ISO)
(R*,R*)-.beta.-[(4-chlorophenyl)methyl]-.alpha.- Bonzi .RTM.
(Syngenta, Retardant (1,1-dimethylethyl)-1H-1,2,4-triazole-1-
Greensboro NC) ethanol (CAS) Difoliant, Ethephon (ANSI)
(2-chloroethyl)phosphonic acid (CAS) Prep .RTM. (Aventis ethylene
CropScience, releaser Research Triangle Park, NC) Gibberellin
Gibberellic acid
(1.alpha.,2.beta.,4a.alpha.,4b.beta.,10.beta.)-2,4a,7-trihydroxy-1-
- RyzUp .RTM. (Valent methyl-8-methylenegibb-3-ene-1,10- U.S.A.
Corp., dicarboxylic acid 1,4a-lactone (CAS) Walnut Creek, CA) Auxin
.alpha.-naphthaleneacetic 1-naphthaleneacetic acid (CAS) Tre-Hold
.RTM. (Amvac acid (ISO) Chemical Co., New Port Beach, CA) Auxin IBA
Indole-3-butyric acid (CAS 8CI) Seradix .RTM. (Aventis CropScience,
Research Triangle Park, NC) Gibberellin BAP + Gibberellic
N-(phenylmethyl)-1H-purine-6-amine and Accel .RTM. (Agtrol acid
gibberellic acid International, Huston, TX)
(S)-trans-2-Amino-4-(2-aminoethoxy)-3- ReTain .RTM. (Valent
butenoic acid hydrochloride U.S.A. Corp., Walnut Creek, CA)
Table 5 is intended as an example. Alternative example product
names and classifications exist for the active ingredients cited
and would fall within the scope of the present invention.
For use in accordance with these methods, suitable hypersensitive
response elicitor protein or polypeptide are from bacterial sources
including, without limitation, Erwinia species (e.g., Erwinia
amylovora, Erwinia chrysanthemi, Erwinia stewartii, Erwinia
carotovora, etc.), Pseudomonas species (e.g., Pseudomonas syringae,
Pseudomonas solanacearum, etc.), and Xanthomonas species (e.g.,
Xanthomonas campestris).
The hypersensitive response elicitor protein or polypeptide is
derived, preferably, from Erwinia chrysanthemi, Erwinia amylovora,
Pseudomonas syringae, Pseudomonas solanacearum, or Xanthomonas
campestris.
A hypersensitive response elicitor protein or polypeptide from
Erwinia chrysanthemi has an amino acid sequence corresponding to
SEQ. ID. No. 1 as follows:
TABLE-US-00006 Met Gln Ile Thr Ile Lys Ala His Ile Gly Gly Asp Leu
Gly Val Ser 1 5 10 15 Gly Leu Gly Ala Gln Gly Leu Lys Gly Leu Asn
Ser Ala Ala Ser Ser 20 25 30 Leu Gly Ser Ser Val Asp Lys Leu Ser
Ser Thr Ile Asp Lys Leu Thr 35 40 45 Ser Ala Leu Thr Ser Met Met
Phe Gly Gly Ala Leu Ala Gln Gly Leu 50 55 60 Gly Ala Ser Ser Lys
Gly Leu Gly Met Ser Asn Gln Leu Gly Gln Ser 65 70 75 80 Phe Gly Asn
Gly Ala Gln Gly Ala Ser Asn Leu Leu Ser Val Pro Lys 85 90 95 Ser
Gly Gly Asp Ala Leu Ser Lys Met Phe Asp Lys Ala Leu Asp Asp 100 105
110 Leu Leu Gly His Asp Thr Val Thr Lys Leu Thr Asn Gln Ser Asn Gln
115 120 125 Leu Ala Asn Ser Met Leu Asn Ala Ser Gln Met Thr Gln Gly
Asn Met 130 135 140 Asn Ala Phe Gly Ser Gly Val Asn Asn Ala Leu Ser
Ser Ile Leu Gly 145 150 155 160 Asn Gly Leu Gly Gln Ser Met Ser Gly
Phe Ser Gln Pro Ser Leu Gly 165 170 175 Ala Gly Gly Leu Gln Gly Leu
Ser Gly Ala Gly Ala Phe Asn Gln Leu 180 185 190 Gly Asn Ala Ile Gly
Met Gly Val Gly Gln Asn Ala Ala Leu Ser Ala 195 200 205 Leu Ser Asn
Val Ser Thr His Val Asp Gly Asn Asn Arg His Phe Val 210 215 220 Asp
Lys Glu Asp Arg Gly Met Ala Lys Glu Ile Gly Gln Phe Met Asp 225 230
235 240 Gln Tyr Pro Glu Ile Phe Gly Lys Pro Glu Tyr Gln Lys Asp Gly
Trp 245 250 255 Ser Ser Pro Lys Thr Asp Asp Lys Ser Trp Ala Lys Ala
Leu Ser Lys 260 265 270 Pro Asp Asp Asp Gly Met Thr Gly Ala Ser Met
Asp Lys Phe Arg Gln 275 280 285 Ala Met Gly Met Ile Lys Ser Ala Val
Ala Gly Asp Thr Gly Asn Thr 290 295 300 Asn Leu Asn Leu Arg Gly Ala
Gly Gly Ala Ser Leu Gly Ile Asp Ala 305 310 315 320 Ala Val Val Gly
Asp Lys Ile Ala Asn Met Ser Leu Gly Lys Leu Ala 325 330 335 Asn
Ala
This hypersensitive response elicitor protein or polypeptide has a
molecular weight of 34 kDa, is heat stable, has a glycine content
of greater than 16%, and contains substantially no cysteine. This
Erwinia chrysanthemi hypersensitive response elicitor protein or
polypeptide is encoded by a DNA molecule having a nucleotide
sequence corresponding to SEQ. ID. No. 2 as follows:
TABLE-US-00007 cgattttacc cgggtgaacg tgctatgacc gacagcatca
cggtattcga caccgttacg 60 gcgtttatgg ccgcgatgaa ccggcatcag
gcggcgcgct ggtcgccgca atccggcgtc 120 gatctggtat ttcagtttgg
ggacaccggg cgtgaactca tgatgcagat tcagccgggg 180 cagcaatatc
ccggcatgtt gcgcacgctg ctcgctcgtc gttatcagca ggcggcagag 240
tgcgatggct gccatctgtg cctgaacggc agcgatgtat tgatcctctg gtggccgctg
300 ccgtcggatc ccggcagtta tccgcaggtg atcgaacgtt tgtttgaact
ggcgggaatg 360 acgttgccgt cgctatccat agcaccgacg gcgcgtccgc
agacagggaa cggacgcgcc 420 cgatcattaa gataaaggcg gcttttttta
ttgcaaaacg gtaacggtga ggaaccgttt 480 caccgtcggc gtcactcagt
aacaagtatc catcatgatg cctacatcgg gatcggcgtg 540 ggcatccgtt
gcagatactt ttgcgaacac ctgacatgaa tgaggaaacg aaattatgca 600
aattacgatc aaagcgcaca tcggcggtga tttgggcgtc tccggtctgg ggctgggtgc
660 tcagggactg aaaggactga attccgcggc ttcatcgctg ggttccagcg
tggataaact 720 gagcagcacc atcgataagt tgacctccgc gctgacttcg
atgatgtttg gcggcgcgct 780 ggcgcagggg ctgggcgcca gctcgaaggg
gctggggatg agcaatcaac tgggccagtc 840 tttcggcaat ggcgcgcagg
gtgcgagcaa cctgctatcc gtaccgaaat ccggcggcga 900 tgcgttgtca
aaaatgtttg ataaagcgct ggacgatctg ctgggtcatg acaccgtgac 960
caagctgact aaccagagca accaactggc taattcaatg ctgaacgcca gccagatgac
1020 ccagggtaat atgaatgcgt tcggcagcgg tgtgaacaac gcactgtcgt
ccattctcgg 1080 caacggtctc ggccagtcga tgagtggctt ctctcagcct
tctctggggg caggcggctt 1140 gcagggcctg agcggcgcgg gtgcattcaa
ccagttgggt aatgccatcg gcatgggcgt 1200 ggggcagaat gctgcgctga
gtgcgttgag taacgtcagc acccacgtag acggtaacaa 1260 ccgccacttt
gtagataaag aagatcgcgg catggcgaaa gagatcggcc agtttatgga 1320
tcagtatccg gaaatattcg gtaaaccgga ataccagaaa gatggctgga gttcgccgaa
1380 gacggacgac aaatcctggg ctaaagcgct gagtaaaccg gatgatgacg
gtatgaccgg 1440 cgccagcatg gacaaattcc gtcaggcgat gggtatgatc
aaaagcgcgg tggcgggtga 1500 taccggcaat accaacctga acctgcgtgg
cgcgggcggt gcatcgctgg gtatcgatgc 1560 ggctgtcgtc ggcgataaaa
tagccaacat gtcgctgggt aagctggcca acgcctgata 1620 atctgtgctg
gcctgataaa gcggaaacga aaaaagagac ggggaagcct gtctcttttc 1680
ttattatgcg gtttatgcgg ttacctggac cggttaatca tcgtcatcga tctggtacaa
1740 acgcacattt tcccgttcat tcgcgtcgtt acgcgccaca atcgcgatgg
catcttcctc 1800 gtcgctcaga ttgcgcggct gatggggaac gccgggtgga
atatagagaa actcgccggc 1860 cagatggaga cacgtctgcg ataaatctgt
gccgtaacgt gtttctatcc gcccctttag 1920 cagatagatt gcggtttcgt
aatcaacatg gtaatgcggt tccgcctgtg cgccggccgg 1980 gatcaccaca
atattcatag aaagctgtct tgcacctacc gtatcgcggg agataccgac 2040
aaaatagggc agtttttgcg tggtatccgt ggggtgttcc ggcctgacaa tcttgagttg
2100 gttcgtcatc atctttctcc atctgggcga cctgatcggt t 2141
The above nucleotide and amino acid sequences are disclosed and
further described in U.S. Pat. No. 5,850,015 to Bauer et al. and
U.S. Pat. No. 5,776,889 to Wei et al., which are hereby
incorporated by reference in their entirety.
One particular hypersensitive response elicitor protein, known as
harpin.sub.Ea, is commercially available from Eden Bioscience
Corporation (Bothell, Wash.) under the name of Messenger.RTM..
Messenger contains 3% by weight of harpin.sub.Ea as the active
ingredient and 97% by weight inert ingredients. Harpin.sub.Ea is
one type of hypersensitive response elicitor protein from Erwinia
amylovora. Harpin.sub.Ea has an amino acid sequence corresponding
to SEQ. ID. No. 3 as follows:
TABLE-US-00008 Met Ser Leu Asn Thr Ser Gly Leu Gly Ala Ser Thr Met
Gln Ile Ser 1 5 10 15 Ile Gly Gly Ala Gly Gly Asn Asn Gly Leu Leu
Gly Thr Ser Arg Gln 20 25 30 Asn Ala Gly Leu Gly Gly Asn Ser Ala
Leu Gly Leu Gly Gly Gly Asn 35 40 45 Gln Asn Asp Thr Val Asn Gln
Leu Ala Gly Leu Leu Thr Gly Met Met 50 55 60 Met Met Met Ser Met
Met Gly Gly Gly Gly Leu Met Gly Gly Gly Leu 65 70 75 80 Gly Gly Gly
Leu Gly Asn Gly Leu Gly Gly Ser Gly Gly Leu Gly Glu 85 90 95 Gly
Leu Ser Asn Ala Leu Asn Asp Met Leu Gly Gly Ser Leu Asn Thr 100 105
110 Leu Gly Ser Lys Gly Gly Asn Asn Thr Thr Ser Thr Thr Asn Ser Pro
115 120 125 Leu Asp Gln Ala Leu Gly Ile Asn Ser Thr Ser Gln Asn Asp
Asp Ser 130 135 140 Thr Ser Gly Thr Asp Ser Thr Ser Asp Ser Ser Asp
Pro Met Gln Gln 145 150 155 160 Leu Leu Lys Met Phe Ser Glu Ile Met
Gln Ser Leu Phe Gly Asp Gly 165 170 175 Gln Asp Gly Thr Gln Gly Ser
Ser Ser Gly Gly Lys Gln Pro Thr Glu 180 185 190 Gly Glu Gln Asn Ala
Tyr Lys Lys Gly Val Thr Asp Ala Leu Ser Gly 195 200 205 Leu Met Gly
Asn Gly Leu Ser Gln Leu Leu Gly Asn Gly Gly Leu Gly 210 215 220 Gly
Gly Gln Gly Gly Asn Ala Gly Thr Gly Leu Asp Gly Ser Ser Leu 225 230
235 240 Gly Gly Lys Gly Leu Gln Asn Leu Ser Gly Pro Val Asp Tyr Gln
Gln 245 250 255 Leu Gly Asn Ala Val Gly Thr Gly Ile Gly Met Lys Ala
Gly Ile Gln 260 265 270 Ala Leu Asn Asp Ile Gly Thr His Arg His Ser
Ser Thr Arg Ser Phe 275 280 285 Val Asn Lys Gly Asp Arg Ala Met Ala
Lys Glu Ile Gly Gln Phe Met 290 295 300 Asp Gln Tyr Pro Glu Val Phe
Gly Lys Pro Gln Tyr Gln Lys Gly Pro 305 310 315 320 Gly Gln Glu Val
Lys Thr Asp Asp Lys Ser Trp Ala Lys Ala Leu Ser 325 330 335 Lys Pro
Asp Asp Asp Gly Met Thr Pro Ala Ser Met Glu Gln Phe Asn 340 345 350
Lys Ala Lys Gly Met Ile Lys Arg Pro Met Ala Gly Asp Thr Gly Asn 355
360 365 Gly Asn Leu Gln Ala Arg Gly Ala Gly Gly Ser Ser Leu Gly Ile
Asp 370 375 380 Ala Met Met Ala Gly Asp Ala Ile Asn Asn Met Ala Leu
Gly Lys Leu 385 390 395 400 Gly Ala Ala
This hypersensitive response elicitor protein or polypeptide has a
molecular weight of about 39 kDa, has a pI of approximately 4.3,
and is heat stable at 100.degree. C. for at least 10 minutes. This
hypersensitive response elicitor protein or polypeptide has
substantially no cysteine. The hypersensitive response elicitor
protein or polypeptide derived from Erwinia amylovora is more fully
described in Wei, Z-M., et al., "Harpin, Elicitor of the
Hypersensitive Response Produced by the Plant Pathogen Erwinia
amylovora," Science 257:85-88 (1992), which is hereby incorporated
by reference in its entirety. The DNA molecule encoding this
hypersensitive response elicitor protein or polypeptide has a
nucleotide sequence corresponding to SEQ. ID. No. 4 as follows:
TABLE-US-00009 aagcttcggc atggcacgtt tgaccgttgg gtcggcaggg
tacgtttgaa ttattcataa 60 gaggaatacg ttatgagtct gaatacaagt
gggctgggag cgtcaacgat gcaaatttct 120 atcggcggtg cgggcggaaa
taacgggttg ctgggtacca gtcgccagaa tgctgggttg 180 ggtggcaatt
ctgcactggg gctgggcggc ggtaatcaaa atgataccgt caatcagctg 240
gctggcttac tcaccggcat gatgatgatg atgagcatga tgggcggtgg tgggctgatg
300 ggcggtggct taggcggtgg cttaggtaat ggcttgggtg gctcaggtgg
cctgggcgaa 360 ggactgtcga acgcgctgaa cgatatgtta ggcggttcgc
tgaacacgct gggctcgaaa 420 ggcggcaaca ataccacttc aacaacaaat
tccccgctgg accaggcgct gggtattaac 480 tcaacgtccc aaaacgacga
ttccacctcc ggcacagatt ccacctcaga ctccagcgac 540 ccgatgcagc
agctgctgaa gatgttcagc gagataatgc aaagcctgtt tggtgatggg 600
caagatggca cccagggcag ttcctctggg ggcaagcagc cgaccgaagg cgagcagaac
660 gcctataaaa aaggagtcac tgatgcgctg tcgggcctga tgggtaatgg
tctgagccag 720 ctccttggca acgggggact gggaggtggt cagggcggta
atgctggcac gggtcttgac 780 ggttcgtcgc tgggcggcaa agggctgcaa
aacctgagcg ggccggtgga ctaccagcag 840 ttaggtaacg ccgtgggtac
cggtatcggt atgaaagcgg gcattcaggc gctgaatgat 900 atcggtacgc
acaggcacag ttcaacccgt tctttcgtca ataaaggcga tcgggcgatg 960
gcgaaggaaa tcggtcagtt catggaccag tatcctgagg tgtttggcaa gccgcagtac
1020 cagaaaggcc cgggtcagga ggtgaaaacc gatgacaaat catgggcaaa
agcactgagc 1080 aagccagatg acgacggaat gacaccagcc agtatggagc
agttcaacaa agccaagggc 1140 atgatcaaaa ggcccatggc gggtgatacc
ggcaacggca acctgcaggc acgcggtgcc 1200 ggtggttctt cgctgggtat
tgatgccatg atggccggtg atgccattaa caatatggca 1260 cttggcaagc
tgggcgcggc ttaagctt 1288
The above nucleotide and amino acid sequences are disclosed and
further described in U.S. Pat. No. 5,849,868 to Beer et al. and
U.S. Pat. No. 5,776,889 to Wei et al., which are hereby
incorporated by reference in their entirety.
Another hypersensitive response elicitor protein or polypeptide
derived from Erwinia amylovora has an amino acid sequence
corresponding to SEQ. ID. No. 5 as follows:
TABLE-US-00010 Met Ser Ile Leu Thr Leu Asn Asn Asn Thr Ser Ser Ser
Pro Gly Leu 1 5 10 15 Phe Gln Ser Gly Gly Asp Asn Gly Leu Gly Gly
His Asn Ala Asn Ser 20 25 30 Ala Leu Gly Gln Gln Pro Ile Asp Arg
Gln Thr Ile Glu Gln Met Ala 35 40 45 Gln Leu Leu Ala Glu Leu Leu
Lys Ser Leu Leu Ser Pro Gln Ser Gly 50 55 60 Asn Ala Ala Thr Gly
Ala Gly Gly Asn Asp Gln Thr Thr Gly Val Gly 65 70 75 80 Asn Ala Gly
Gly Leu Asn Gly Arg Lys Gly Thr Ala Gly Thr Thr Pro 85 90 95 Gln
Ser Asp Ser Gln Asn Met Leu Ser Glu Met Gly Asn Asn Gly Leu 100 105
110 Asp Gln Ala Ile Thr Pro Asp Gly Gln Gly Gly Gly Gln Ile Gly Asp
115 120 125 Asn Pro Leu Leu Lys Ala Met Leu Lys Leu Ile Ala Arg Met
Met Asp 130 135 140 Gly Gln Ser Asp Gln Phe Gly Gln Pro Gly Thr Gly
Asn Asn Ser Ala 145 150 155 160 Ser Ser Gly Thr Ser Ser Ser Gly Gly
Ser Pro Phe Asn Asp Leu Ser 165 170 175 Gly Gly Lys Ala Pro Ser Gly
Asn Ser Pro Ser Gly Asn Tyr Ser Pro 180 185 190 Val Ser Thr Phe Ser
Pro Pro Ser Thr Pro Thr Ser Pro Thr Ser Pro 195 200 205 Leu Asp Phe
Pro Ser Ser Pro Thr Lys Ala Ala Gly Gly Ser Thr Pro 210 215 220 Val
Thr Asp His Pro Asp Pro Val Gly Ser Ala Gly Ile Gly Ala Gly 225 230
235 240 Asn Ser Val Ala Phe Thr Ser Ala Gly Ala Asn Gln Thr Val Leu
His 245 250 255 Asp Thr Ile Thr Val Lys Ala Gly Gln Val Phe Asp Gly
Lys Gly Gln 260 265 270 Thr Phe Thr Ala Gly Ser Glu Leu Gly Asp Gly
Gly Gln Ser Glu Asn 275 280 285 Gln Lys Pro Leu Phe Ile Leu Glu Asp
Gly Ala Ser Leu Lys Asn Val 290 295 300 Thr Met Gly Asp Asp Gly Ala
Asp Gly Ile His Leu Tyr Gly Asp Ala 305 310 315 320 Lys Ile Asp Asn
Leu His Val Thr Asn Val Gly Glu Asp Ala Ile Thr 325 330 335 Val Lys
Pro Asn Ser Ala Gly Lys Lys Ser His Val Glu Ile Thr Asn 340 345 350
Ser Ser Phe Glu His Ala Ser Asp Lys Ile Leu Gln Leu Asn Ala Asp 355
360 365 Thr Asn Leu Ser Val Asp Asn Val Lys Ala Lys Asp Phe Gly Thr
Phe 370 375 380 Val Arg Thr Asn Gly Gly Gln Gln Gly Asn Trp Asp Leu
Asn Leu Ser 385 390 395 400 His Ile Ser Ala Glu Asp Gly Lys Phe Ser
Phe Val Lys Ser Asp Ser 405 410 415 Glu Gly Leu Asn Val Asn Thr Ser
Asp Ile Ser Leu Gly Asp Val Glu 420 425 430 Asn His Tyr Lys Val Pro
Met Ser Ala Asn Leu Lys Val Ala Glu 435 440 445
This protein or polypeptide is acidic, rich in glycine and serine,
and lacks cysteine. It is also heat stable, protease sensitive, and
suppressed by inhibitors of plant metabolism. The protein or
polypeptide of the present invention has a predicted molecular size
of ca. 45 kDa. The DNA molecule encoding this hypersensitive
response elicitor protein or polypeptide has a nucleotide sequence
corresponding to SEQ. ID. No. 6 as follows:
TABLE-US-00011 atgtcaattc ttacgcttaa caacaatacc tcgtcctcgc
cgggtctgtt ccagtccggg 60 ggggacaacg ggcttggtgg tcataatgca
aattctgcgt tggggcaaca acccatcgat 120 cggcaaacca ttgagcaaat
ggctcaatta ttggcggaac tgttaaagtc actgctatcg 180 ccacaatcag
gtaatgcggc aaccggagcc ggtggcaatg accagactac aggagttggt 240
aacgctggcg gcctgaacgg acgaaaaggc acagcaggaa ccactccgca gtctgacagt
300 cagaacatgc tgagtgagat gggcaacaac gggctggatc aggccatcac
gcccgatggc 360 cagggcggcg ggcagatcgg cgataatcct ttactgaaag
ccatgctgaa gcttattgca 420 cgcatgatgg acggccaaag cgatcagttt
ggccaacctg gtacgggcaa caacagtgcc 480 tcttccggta cttcttcatc
tggcggttcc ccttttaacg atctatcagg ggggaaggcc 540 ccttccggca
actccccttc cggcaactac tctcccgtca gtaccttctc acccccatcc 600
acgccaacgt cccctacctc accgcttgat ttcccttctt ctcccaccaa agcagccggg
660 ggcagcacgc cggtaaccga tcatcctgac cctgttggta gcgcgggcat
cggggccgga 720 aattcggtgg ccttcaccag cgccggcgct aatcagacgg
tgctgcatga caccattacc 780 gtgaaagcgg gtcaggtgtt tgatggcaaa
ggacaaacct tcaccgccgg ttcagaatta 840 ggcgatggcg gccagtctga
aaaccagaaa ccgctgttta tactggaaga cggtgccagc 900 ctgaaaaacg
tcaccatggg cgacgacggg gcggatggta ttcatcttta cggtgatgcc 960
aaaatagaca atctgcacgt caccaacgtg ggtgaggacg cgattaccgt taagccaaac
1020 agcgcgggca aaaaatccca cgttgaaatc actaacagtt ccttcgagca
cgcctctgac 1080 aagatcctgc agctgaatgc cgatactaac ctgagcgttg
acaacgtgaa ggccaaagac 1140 tttggtactt ttgtacgcac taacggcggt
caacagggta actgggatct gaatctgagc 1200 catatcagcg cagaagacgg
taagttctcg ttcgttaaaa gcgatagcga ggggctaaac 1260 gtcaatacca
gtgatatctc actgggtgat gttgaaaacc actacaaagt gccgatgtcc 1320
gccaacctga aggtggctga atga 1344
The above nucleotide and amino acid sequences are disclosed and
further described in PCT Application Publication No. WO 99/07208 to
Kim et al., which is hereby incorporated by reference in its
entirety.
A hypersensitive response elicitor protein or polypeptide derived
from Pseudomonas syringae has an amino acid sequence corresponding
to SEQ. ID. No. 7 as follows:
TABLE-US-00012 Met Gln Ser Leu Ser Leu Asn Ser Ser Ser Leu Gln Thr
Pro Ala Met 1 5 10 15 Ala Leu Val Leu Val Arg Pro Glu Ala Glu Thr
Thr Gly Ser Thr Ser 20 25 30 Ser Lys Ala Leu Gln Glu Val Val Val
Lys Leu Ala Glu Glu Leu Met 35 40 45 Arg Asn Gly Gln Leu Asp Asp
Ser Ser Pro Leu Gly Lys Leu Leu Ala 50 55 60 Lys Ser Met Ala Ala
Asp Gly Lys Ala Gly Gly Gly Ile Glu Asp Val 65 70 75 80 Ile Ala Ala
Leu Asp Lys Leu Ile His Glu Lys Leu Gly Asp Asn Phe 85 90 95 Gly
Ala Ser Ala Asp Ser Ala Ser Gly Thr Gly Gln Gln Asp Leu Met 100 105
110 Thr Gln Val Leu Asn Gly Leu Ala Lys Ser Met Leu Asp Asp Leu Leu
115 120 125 Thr Lys Gln Asp Gly Gly Thr Ser Phe Ser Glu Asp Asp Met
Pro Met 130 135 140 Leu Asn Lys Ile Ala Gln Phe Met Asp Asp Asn Pro
Ala Gln Phe Pro 145 150 155 160 Lys Pro Asp Ser Gly Ser Trp Val Asn
Glu Leu Lys Glu Asp Asn Phe 165 170 175 Leu Asp Gly Asp Glu Thr Ala
Ala Phe Arg Ser Ala Leu Asp Ile Ile 180 185 190 Gly Gln Gln Leu Gly
Asn Gln Gln Ser Asp Ala Gly Ser Leu Ala Gly 195 200 205 Thr Gly Gly
Gly Leu Gly Thr Pro Ser Ser Phe Ser Asn Asn Ser Ser 210 215 220 Val
Met Gly Asp Pro Leu Ile Asp Ala Asn Thr Gly Pro Gly Asp Ser 225 230
235 240 Gly Asn Thr Arg Gly Glu Ala Gly Gln Leu Ile Gly Glu Leu Ile
Asp 245 250 255 Arg Gly Leu Gln Ser Val Leu Ala Gly Gly Gly Leu Gly
Thr Pro Val 260 265 270 Asn Thr Pro Gln Thr Gly Thr Ser Ala Asn Gly
Gly Gln Ser Ala Gln 275 280 285 Asp Leu Asp Gln Leu Leu Gly Gly Leu
Leu Leu Lys Gly Leu Glu Ala 290 295 300 Thr Leu Lys Asp Ala Gly Gln
Thr Gly Thr Asp Val Gln Ser Ser Ala 305 310 315 320 Ala Gln Ile Ala
Thr Leu Leu Val Ser Thr Leu Leu Gln Gly Thr Arg 325 330 335 Asn Gln
Ala Ala Ala 340
This hypersensitive response elicitor protein or polypeptide has a
molecular weight of 34-35 kDa. It is rich in glycine (about 13.5%)
and lacks cysteine and tyrosine. Further information about the
hypersensitive response elicitor derived from Pseudomonas syringae
is found in He, S. Y., et al., "Pseudomonas syringae pv. syringae
Harpin.sub.Pss: a Protein that is Secreted via the Hrp Pathway and
Elicits the Hypersensitive Response in Plants," Cell 73:1255-1266
(1993), which is hereby incorporated by reference in its entirety.
The DNA molecule encoding this hypersensitive response elicitor
from Pseudomonas syringae has a nucleotide sequence corresponding
to SEQ. ID. No. 8 as follows:
TABLE-US-00013 atgcagagtc tcagtcttaa cagcagctcg ctgcaaaccc
cggcaatggc ccttgtcctg 60 gtacgtcctg aagccgagac gactggcagt
acgtcgagca aggcgcttca ggaagttgtc 120 gtgaagctgg ccgaggaact
gatgcgcaat ggtcaactcg acgacagctc gccattggga 180 aaactgttgg
ccaagtcgat ggccgcagat ggcaaggcgg gcggcggtat tgaggatgtc 240
atcgctgcgc tggacaagct gatccatgaa aagctcggtg acaacttcgg cgcgtctgcg
300 gacagcgcct cgggtaccgg acagcaggac ctgatgactc aggtgctcaa
tggcctggcc 360 aagtcgatgc tcgatgatct tctgaccaag caggatggcg
ggacaagctt ctccgaagac 420 gatatgccga tgctgaacaa gatcgcgcag
ttcatggatg acaatcccgc acagtttccc 480 aagccggact cgggctcctg
ggtgaacgaa ctcaaggaag acaacttcct tgatggcgac 540 gaaacggctg
cgttccgttc ggcactcgac atcattggcc agcaactggg taatcagcag 600
agtgacgctg gcagtctggc agggacgggt ggaggtctgg gcactccgag cagtttttcc
660 aacaactcgt ccgtgatggg tgatccgctg atcgacgcca ataccggtcc
cggtgacagc 720 ggcaataccc gtggtgaagc ggggcaactg atcggcgagc
ttatcgaccg tggcctgcaa 780 tcggtattgg ccggtggtgg actgggcaca
cccgtaaaca ccccgcagac cggtacgtcg 840 gcgaatggcg gacagtccgc
tcaggatctt gatcagttgc tgggcggctt gctgctcaag 900 ggcctggagg
caacgctcaa ggatgccggg caaacaggca ccgacgtgca gtcgagcgct 960
gcgcaaatcg ccaccttgct ggtcagtacg ctgctgcaag gcacccgcaa tcaggctgca
1020 gcctga 1026
The above nucleotide and amino acid sequences are disclosed and
further described in U.S. Pat. No. 5,708,139 to Collmer et al. and
U.S. Pat. No. 5,776,889 to Wei et al., which are hereby
incorporated by reference in their entirety.
Another hypersensitive response elicitor protein or polypeptide
derived from Pseudomonas syringae has an amino acid sequence
corresponding to SEQ. ID. No. 9 as follows:
TABLE-US-00014 Met Ser Ile Gly Ile Thr Pro Arg Pro Gln Gln Thr Thr
Thr Pro Leu 1 5 10 15 Asp Phe Ser Ala Leu Ser Gly Lys Ser Pro Gln
Pro Asn Thr Phe Gly 20 25 30 Glu Gln Asn Thr Gln Gln Ala Ile Asp
Pro Ser Ala Leu Leu Phe Gly 35 40 45 Ser Asp Thr Gln Lys Asp Val
Asn Phe Gly Thr Pro Asp Ser Thr Val 50 55 60 Gln Asn Pro Gln Asp
Ala Ser Lys Pro Asn Asp Ser Gln Ser Asn Ile 65 70 75 80 Ala Lys Leu
Ile Ser Ala Leu Ile Met Ser Leu Leu Gln Met Leu Thr 85 90 95 Asn
Ser Asn Lys Lys Gln Asp Thr Asn Gln Glu Gln Pro Asp Ser Gln 100 105
110 Ala Pro Phe Gln Asn Asn Gly Gly Leu Gly Thr Pro Ser Ala Asp Ser
115 120 125 Gly Gly Gly Gly Thr Pro Asp Ala Thr Gly Gly Gly Gly Gly
Asp Thr 130 135 140 Pro Ser Ala Thr Gly Gly Gly Gly Gly Asp Thr Pro
Thr Ala Thr Gly 145 150 155 160 Gly Gly Gly Ser Gly Gly Gly Gly Thr
Pro Thr Ala Thr Gly Gly Gly 165 170 175 Ser Gly Gly Thr Pro Thr Ala
Thr Gly Gly Gly Glu Gly Gly Val Thr 180 185 190 Pro Gln Ile Thr Pro
Gln Leu Ala Asn Pro Asn Arg Thr Ser Gly Thr 195 200 205 Gly Ser Val
Ser Asp Thr Ala Gly Ser Thr Glu Gln Ala Gly Lys Ile 210 215 220 Asn
Val Val Lys Asp Thr Ile Lys Val Gly Ala Gly Glu Val Phe Asp 225 230
235 240 Gly His Gly Ala Thr Phe Thr Ala Asp Lys Ser Met Gly Asn Gly
Asp 245 250 255 Gln Gly Glu Asn Gln Lys Pro Met Phe Glu Leu Ala Glu
Gly Ala Thr 260 265 270 Leu Lys Asn Val Asn Leu Gly Glu Asn Glu Val
Asp Gly Ile His Val 275 280 285 Lys Ala Lys Asn Ala Gln Glu Val Thr
Ile Asp Asn Val His Ala Gln 290 295 300 Asn Val Gly Glu Asp Leu Ile
Thr Val Lys Gly Glu Gly Gly Ala Ala 305 310 315 320 Val Thr Asn Leu
Asn Ile Lys Asn Ser Ser Ala Lys Gly Ala Asp Asp 325 330 335 Lys Val
Val Gln Leu Asn Ala Asn Thr His Leu Lys Ile Asp Asn Phe 340 345 350
Lys Ala Asp Asp Phe Gly Thr Met Val Arg Thr Asn Gly Gly Lys Gln 355
360 365 Phe Asp Asp Met Ser Ile Glu Leu Asn Gly Ile Glu Ala Asn His
Gly 370 375 380 Lys Phe Ala Leu Val Lys Ser Asp Ser Asp Asp Leu Lys
Leu Ala Thr 385 390 395 400 Gly Asn Ile Ala Met Thr Asp Val Lys His
Ala Tyr Asp Lys Thr Gln 405 410 415 Ala Ser Thr Gln His Thr Glu Leu
420
This protein or polypeptide is acidic, glycine-rich, lacks
cysteine, and is deficient in aromatic amino acids. The DNA
molecule encoding this hypersensitive response elicitor from
Pseudomonas syringae has a nucleotide sequence corresponding to
SEQ. ID. No. 10 as follows:
TABLE-US-00015 tccacttcgc tgattttgaa attggcagat tcatagaaac
gttcaggtgt ggaaatcagg 60 ctgagtgcgc agatttcgtt gataagggtg
tggtactggt cattgttggt catttcaagg 120 cctctgagtg cggtgcggag
caataccagt cttcctgctg gcgtgtgcac actgagtcgc 180 aggcataggc
atttcagttc cttgcgttgg ttgggcatat aaaaaaagga acttttaaaa 240
acagtgcaat gagatgccgg caaaacggga accggtcgct gcgctttgcc actcacttcg
300 agcaagctca accccaaaca tccacatccc tatcgaacgg acagcgatac
ggccacttgc 360 tctggtaaac cctggagctg gcgtcggtcc aattgcccac
ttagcgaggt aacgcagcat 420 gagcatcggc atcacacccc ggccgcaaca
gaccaccacg ccactcgatt tttcggcgct 480 aagcggcaag agtcctcaac
caaacacgtt cggcgagcag aacactcagc aagcgatcga 540 cccgagtgca
ctgttgttcg gcagcgacac acagaaagac gtcaacttcg gcacgcccga 600
cagcaccgtc cagaatccgc aggacgccag caagcccaac gacagccagt ccaacatcgc
660 taaattgatc agtgcattga tcatgtcgtt gctgcagatg ctcaccaact
ccaataaaaa 720 gcaggacacc aatcaggaac agcctgatag ccaggctcct
ttccagaaca acggcgggct 780 cggtacaccg tcggccgata gcgggggcgg
cggtacaccg gatgcgacag gtggcggcgg 840 cggtgatacg ccaagcgcaa
caggcggtgg cggcggtgat actccgaccg caacaggcgg 900 tggcggcagc
ggtggcggcg gcacacccac tgcaacaggt ggcggcagcg gtggcacacc 960
cactgcaaca ggcggtggcg agggtggcgt aacaccgcaa atcactccgc agttggccaa
1020 ccctaaccgt acctcaggta ctggctcggt gtcggacacc gcaggttcta
ccgagcaagc 1080 cggcaagatc aatgtggtga aagacaccat caaggtcggc
gctggcgaag tctttgacgg 1140 ccacggcgca accttcactg ccgacaaatc
tatgggtaac ggagaccagg gcgaaaatca 1200 gaagcccatg ttcgagctgg
ctgaaggcgc tacgttgaag aatgtgaacc tgggtgagaa 1260 cgaggtcgat
ggcatccacg tgaaagccaa aaacgctcag gaagtcacca ttgacaacgt 1320
gcatgcccag aacgtcggtg aagacctgat tacggtcaaa ggcgagggag gcgcagcggt
1380 cactaatctg aacatcaaga acagcagtgc caaaggtgca gacgacaagg
ttgtccagct 1440 caacgccaac actcacttga aaatcgacaa cttcaaggcc
gacgatttcg gcacgatggt 1500 tcgcaccaac ggtggcaagc agtttgatga
catgagcatc gagctgaacg gcatcgaagc 1560 taaccacggc aagttcgccc
tggtgaaaag cgacagtgac gatctgaagc tggcaacggg 1620 caacatcgcc
atgaccgacg tcaaacacgc ctacgataaa acccaggcat cgacccaaca 1680
caccgagctt tgaatccaga caagtagctt gaaaaaaggg ggtggactc 1729
The above nucleotide and amino acid sequences are disclosed and
further described in U.S. Pat. No. 6,172,184 to Collmer et al.,
which is hereby incorporated by reference in its entirety.
A hypersensitive response elicitor protein or polypeptide derived
from Pseudomonas solanacearum has an amino acid sequence
corresponding to SEQ. ID. No. 11 as follows:
TABLE-US-00016 Met Ser Val Gly Asn Ile Gln Ser Pro Ser Asn Leu Pro
Gly Leu Gln 1 5 10 15 Asn Leu Asn Leu Asn Thr Asn Thr Asn Ser Gln
Gln Ser Gly Gln Ser 20 25 30 Val Gln Asp Leu Ile Lys Gln Val Glu
Lys Asp Ile Leu Asn Ile Ile 35 40 45 Ala Ala Leu Val Gln Lys Ala
Ala Gln Ser Ala Gly Gly Asn Thr Gly 50 55 60 Asn Thr Gly Asn Ala
Pro Ala Lys Asp Gly Asn Ala Asn Ala Gly Ala 65 70 75 80 Asn Asp Pro
Ser Lys Asn Asp Pro Ser Lys Ser Gln Ala Pro Gln Ser 85 90 95 Ala
Asn Lys Thr Gly Asn Val Asp Asp Ala Asn Asn Gln Asp Pro Met 100 105
110 Gln Ala Leu Met Gln Leu Leu Glu Asp Leu Val Lys Leu Leu Lys Ala
115 120 125 Ala Leu His Met Gln Gln Pro Gly Gly Asn Asp Lys Gly Asn
Gly Val 130 135 140 Gly Gly Ala Asn Gly Ala Lys Gly Ala Gly Gly Gln
Gly Gly Leu Ala 145 150 155 160 Glu Ala Leu Gln Glu Ile Glu Gln Ile
Leu Ala Gln Leu Gly Gly Gly 165 170 175 Gly Ala Gly Ala Gly Gly Ala
Gly Gly Gly Val Gly Gly Ala Gly Gly 180 185 190 Ala Asp Gly Gly Ser
Gly Ala Gly Gly Ala Gly Gly Ala Asn Gly Ala 195 200 205 Asp Gly Gly
Asn Gly Val Asn Gly Asn Gln Ala Asn Gly Pro Gln Asn 210 215 220 Ala
Gly Asp Val Asn Gly Ala Asn Gly Ala Asp Asp Gly Ser Glu Asp 225 230
235 240 Gln Gly Gly Leu Thr Gly Val Leu Gln Lys Leu Met Lys Ile Leu
Asn 245 250 255 Ala Leu Val Gln Met Met Gln Gln Gly Gly Leu Gly Gly
Gly Asn Gln 260 265 270 Ala Gln Gly Gly Ser Lys Gly Ala Gly Asn Ala
Ser Pro Ala Ser Gly 275 280 285 Ala Asn Pro Gly Ala Asn Gln Pro Gly
Ser Ala Asp Asp Gln Ser Ser 290 295 300 Gly Gln Asn Asn Leu Gln Ser
Gln Ile Met Asp Val Val Lys Glu Val 305 310 315 320 Val Gln Ile Leu
Gln Gln Met Leu Ala Ala Gln Asn Gly Gly Ser Gln 325 330 335 Gln Ser
Thr Ser Thr Gln Pro Met 340
Further information regarding this hypersensitive response elicitor
protein or polypeptide derived from Pseudomonas solanacearum is set
forth in Arlat, M., et al., "PopA1, a Protein which Induces a
Hypersensitive-like Response in Specific Petunia Genotypes, is
Secreted via the Hrp Pathway of Pseudomonas solanacearum," EMBO J.
13:543-533 (1994), which is hereby incorporated by reference in its
entirety. It is encoded by a DNA molecule from Pseudomonas
solanacearum having a nucleotide sequence corresponding SEQ. ID.
No. 12 as follows:
TABLE-US-00017 atgtcagtcg gaaacatcca gagcccgtcg aacctcccgg
gtctgcagaa cctgaacctc 60 aacaccaaca ccaacagcca gcaatcgggc
cagtccgtgc aagacctgat caagcaggtc 120 gagaaggaca tcctcaacat
catcgcagcc ctcgtgcaga aggccgcaca gtcggcgggc 180 ggcaacaccg
gtaacaccgg caacgcgccg gcgaaggacg gcaatgccaa cgcgggcgcc 240
aacgacccga gcaagaacga cccgagcaag agccaggctc cgcagtcggc caacaagacc
300 ggcaacgtcg acgacgccaa caaccaggat ccgatgcaag cgctgatgca
gctgctggaa 360 gacctggtga agctgctgaa ggcggccctg cacatgcagc
agcccggcgg caatgacaag 420 ggcaacggcg tgggcggtgc caacggcgcc
aagggtgccg gcggccaggg cggcctggcc 480 gaagcgctgc aggagatcga
gcagatcctc gcccagctcg gcggcggcgg tgctggcgcc 540 ggcggcgcgg
gtggcggtgt cggcggtgct ggtggcgcgg atggcggctc cggtgcgggt 600
ggcgcaggcg gtgcgaacgg cgccgacggc ggcaatggcg tgaacggcaa ccaggcgaac
660 ggcccgcaga acgcaggcga tgtcaacggt gccaacggcg cggatgacgg
cagcgaagac 720 cagggcggcc tcaccggcgt gctgcaaaag ctgatgaaga
tcctgaacgc gctggtgcag 780 atgatgcagc aaggcggcct cggcggcggc
aaccaggcgc agggcggctc gaagggtgcc 840 ggcaacgcct cgccggcttc
cggcgcgaac ccgggcgcga accagcccgg ttcggcggat 900 gatcaatcgt
ccggccagaa caatctgcaa tcccagatca tggatgtggt gaaggaggtc 960
gtccagatcc tgcagcagat gctggcggcg cagaacggcg gcagccagca gtccacctcg
1020 acgcagccga tgtaa 1035
The above nucleotide and amino acid sequences are disclosed and
further described in U.S. Pat. No. 5,776,889 to Wei et al., which
is hereby incorporated by reference in its entirety.
A hypersensitive response elicitor polypeptide or protein derived
from Xanthomonas campestris has an amino acid sequence
corresponding to SEQ. ID. No. 13 as follows:
TABLE-US-00018 Met Asp Ser Ile Gly Asn Asn Phe Ser Asn Ile Gly Asn
Leu Gln Thr 1 5 10 15 Met Gly Ile Gly Pro Gln Gln His Glu Asp Ser
Ser Gln Gln Ser Pro 20 25 30 Ser Ala Gly Ser Glu Gln Gln Leu Asp
Gln Leu Leu Ala Met Phe Ile 35 40 45 Met Met Met Leu Gln Gln Ser
Gln Gly Ser Asp Ala Asn Gln Glu Cys 50 55 60 Gly Asn Glu Gln Pro
Gln Asn Gly Gln Gln Glu Gly Leu Ser Pro Leu 65 70 75 80 Thr Gln Met
Leu Met Gln Ile Val Met Gln Leu Met Gln Asn Gln Gly 85 90 95 Gly
Ala Gly Met Gly Gly Gly Gly Ser Val Asn Ser Ser Leu Gly Gly 100 105
110 Asn Ala
This hypersensitive response elicitor polypeptide or protein has an
estimated molecular weight of about 12 kDa based on the deduced
amino acid sequence, which is consistent with a molecular weight of
about 14 kDa as detected by SDS-PAGE. The above protein or
polypeptide is encoded by a DNA molecule according to SEQ. ID. No.
14 as follows:
TABLE-US-00019 atggactcta tcggaaacaa cttttcgaat atcggcaacc
tgcagacgat gggcatcggg 60 cctcagcaac acgaggactc cagccagcag
tcgccttcgg ctggctccga gcagcagctg 120 gatcagttgc tcgccatgtt
catcatgatg atgctgcaac agagccaggg cagcgatgca 180 aatcaggagt
gtggcaacga acaaccgcag aacggtcaac aggaaggcct gagtccgttg 240
acgcagatgc tgatgcagat cgtgatgcag ctgatgcaga accagggcgg cgccggcatg
300 ggcggtggcg gttcggtcaa cagcagcctg ggcggcaacg cc 342
The above nucleotide and amino acid sequences are disclosed and
further described in U.S. patent application Ser. No. 09/829,124,
which is hereby incorporated by reference in its entirety.
Other embodiments of the present invention include, but are not
limited to, use of a hypersensitive response elicitor protein or
polypeptide derived from Erwinia carotovora and Erwinia stewartii.
Isolation of Erwinia carotovora hypersensitive response elicitor
protein or polypeptide is described in Cui, et al., "The RsmA
Mutants of Erwinia carotovora subsp. carotovora Strain Ecc7
Overexpress hrp N.sub.Ecc and Elicit a Hypersensitive Reaction-like
Response in Tobacco Leaves," MPMI, 9(7):565-73 (1996), which is
hereby incorporated by reference in its entirety. A hypersensitive
response elicitor protein or polypeptide of Erwinia stewartii is
set forth in Ahmad, et al., "Harpin is Not Necessary for the
Pathogenicity of Erwinia stewartii on Maize," 8th Int'l. Cone.
Molec. Plant-Microbe Interact., Jul. 14-19, 1996 and Ahmad, et al.,
"Harpin is Not Necessary for the Pathogenicity of Erwinia stewartii
on Maize," Ann. Mtg. Am. Phytopath. Soc., Jul. 27-31, 1996, which
are hereby incorporated by reference in their entirety.
Other elicitors can be readily identified by isolating putative
hypersensitive response elicitors and testing them for elicitor
activity as described, for example, in Wei, Z-M., et al., "Harpin,
Elicitor of the Hypersensitive Response Produced by the Plant
Pathogen Erwinia amylovora," Science 257:85-88 (1992), which is
hereby incorporated by reference in its entirety. Cell-free
preparations from culture supernatants can be tested for elicitor
activity (i.e., local necrosis) by using them to infiltrate
appropriate plant tissues. Once identified, DNA molecules encoding
a hypersensitive response elicitor can be isolated using standard
techniques known to those skilled in the art.
The hypersensitive response elicitor protein or polypeptide can
also be a fragment of the above referenced hypersensitive response
elicitor proteins or polypeptides as well as fragments of full
length elicitors from other pathogens.
Suitable fragments can be produced by several means. Subclones of
the gene encoding a known elicitor protein can be produced using
conventional molecular genetic manipulation for subcloning gene
fragments, such as described by Sambrook et al., Molecular Cloning:
A Laboratory Manual, Cold Springs Laboratory, Cold Springs Harbor,
N.Y. (1989), and Ausubel et al. (ed.), Current Protocols in
Molecular Biology, John Wiley & Sons (New York, N.Y.) (1999 and
preceding editions), which are hereby incorporated by reference in
their entirety. The subolones then are expressed in vitro or in
vivo in bacterial cells to yield a smaller protein or polypeptide
that can be tested for elicitor activity, e.g., using procedures
set forth in Wei, Z-M., et al., Science 257: 85-88 (1992), which is
hereby incorporated by reference in its entirety.
In another approach, based on knowledge of the primary structure of
the protein, fragments of the elicitor protein gene may be
synthesized using the PCR technique together with specific sets of
primers chosen to represent particular portions of the protein.
Erlich, H. A., et al., "Recent Advances in the Polymerase Chain
Reaction," Science 252:1643-51 (1991), which is hereby incorporated
by reference in its entirety. These can then be cloned into an
appropriate vector for expression of a truncated protein or
polypeptide from bacterial cells as described above.
Examples of suitable fragments of a hypersensitive response
elicitor are described in WIPO International Publication Numbers:
WO 98/54214 and WO 01/98501, which are hereby incorporated by
reference in their entirety.
DNA molecules encoding a hypersensitive response elicitor protein
or polypeptide can also include a DNA molecule that hybridizes
under stringent conditions to the DNA molecule having a nucleotide
sequences from one of the above identified hypersensitive response
licitors. An example of suitable stringency conditions is when
hybridization is carried out at a temperature of about 37.degree.
C. using a hybridization medium that includes 0.9M sodium citrate
("SSC") buffer, followed by washing with 0.2.times.SSC buffer at
37.degree. C. Higher stringency can readily be attained by
increasing the temperature for either hybridization or washing
conditions or increasing the sodium concentration of the
hybridization or wash medium. Nonspecific binding may also be
controlled using any one of a number of known techniques such as,
for example, blocking the membrane with protein-containing
solutions, addition of heterologous RNA, DNA, and SDS to the
hybridization buffer, and treatment with RNase. Wash conditions are
typically performed at or below stringency. Exemplary high
stringency conditions include carrying out hybridization at a
temperature of about 42.degree. C. to about 65.degree. C. for up to
about 20 hours in a hybridization medium containing 1M NaCl, 50 mM
Tris-HCl, pH 7.4, 10 mM EDTA, 0.1% sodium dodecyl sulfate (SDS),
0.2% ficoll, 0.2% polyvinylpyrrolidone, 0.2% bovine serum albumin,
and 50 .mu.g/ml E. coli DNA, followed by washing carried out at
between about 42.degree. C. to about 65.degree. C. in a
0.2.times.SSC buffer.
Variants of suitable hypersensitive response elicitor proteins or
polypeptides can also be expressed. Variants may be made by, for
example, the deletion, addition, or alteration of amino acids that
have minimal influence on the properties, secondary structure and
hydropathic nature of the polypeptide. For example, a polypeptide
may be conjugated to a signal (or leader) sequence at the
N-terminal end of the protein which co-translationally or
post-translationally directs transfer of the protein. The
polypeptide may also be conjugated to a linker or other sequence
for ease of synthesis, purification, or identification of the
polypeptide.
The DNA molecule encoding the hypersensitive response elicitor
polypeptide or protein can be incorporated in cells using
conventional recombinant DNA technology. Generally, this involves
inserting the DNA molecule into an expression system to which the
DNA molecule is heterologous (i.e. not normally present). The
heterologous DNA molecule is inserted into the expression system or
vector in sense orientation and correct reading frame. The vector
contains the necessary elements for the transcription and
translation of the inserted protein-coding sequences.
U.S. Pat. No. 4,237,224 to Cohen and Boyer, which is hereby
incorporated by reference in its entirety, describes the production
of expression systems in the form of recombinant plasmids using
restriction enzyme cleavage and ligation with DNA ligase. These
recombinant plasmids are then introduced by means of transformation
and replicated in unicellular cultures including procaryotic
organisms and eucaryotic cells grown in tissue culture.
Recombinant genes may also be introduced into viruses, such as
vaccina virus. Recombinant viruses can be generated by transfection
of plasmids into cells infected with virus.
Suitable vectors include, but are not limited to, the following
viral vectors such as lambda vector system gt11, gt WES.tB, Charon
4, and plasmid vectors such as pBR322, pBR325, pACYC177, pACYC1084,
pUC8, pUC9, pUC18, pUC19, pLG339, pR290, pKC37, pKC101, SV 40,
pBluescript II SK +/- or KS +/- (see "Stratagene Cloning Systems"
Catalog (1993) from Stratagene, La Jolla, Calif., which is hereby
incorporated by reference in its entirety), pQE, pIH821, pGEX, pET
series (see F. W. Studier et. al., "Use of T7 RNA Polymerase to
Direct Expression of Cloned Genes," Gene Expression Technology vol.
185 (1990), which is hereby incorporated by reference in its
entirety), and any derivatives thereof. Recombinant molecules can
be introduced into cells via transformation, particularly
transduction, conjugation, mobilization, or electroporation. The
DNA sequences are cloned into the vector using standard cloning
procedures in the art, as described by Sambrook et al., Molecular
Cloning: A Laboratory Manual, Cold Springs Laboratory, Cold Springs
Harbor, N.Y. (1989), which is hereby incorporated by reference in
its entirety.
A variety of host-vector systems may be utilized to express the
protein-encoding sequence(s). Primarily, the vector system must be
compatible with the host cell used. Host-vector systems include but
are not limited to the following: bacteria transformed with
bacteriophage DNA, plasmid DNA, or cosmid DNA; microorganisms such
as yeast containing yeast vectors; mammalian cell systems infected
with virus (e.g., vaccinia virus, adenovirus, etc.); insect cell
systems infected with virus (e.g., baculovirus); and plant cells
infected by bacteria. The expression elements of these vectors vary
in their strength and specificities. Depending upon the host-vector
system utilized, any one of a number of suitable transcription and
translation elements can be used.
Different genetic signals and processing events control many levels
of gene expression (e.g., DNA transcription and messenger RNA
(mRNA) translation).
Transcription of DNA is dependent upon the presence of a promotor
which is a DNA sequence that directs the binding of RNA polymerase
and thereby promotes mRNA synthesis. The DNA sequences of
eucaryotic promotors differ from those of procaryotic promotors.
Furthermore, eucaryotic promotors and accompanying genetic signals
may not be recognized in or may not function in a procaryotic
system, and, further, procaryotic promotors are not recognized and
do not function in eucaryotic cells.
Similarly, translation of mRNA in prokaryotes depends upon the
presence of the proper prokaryotic signals which differ from those
of eukaryotes. Efficient translation of mRNA in prokaryotes
requires a ribosome binding site called the Shine-Dalgarno ("SD")
sequence on the mRNA. This sequence is a short nucleotide sequence
of MRNA that is located before the start codon, usually AUG, which
encodes the amino-terminal methionine of the protein. The SD
sequences are complementary to the 3'-end of the 16S rRNA
(ribosomal RNA) and probably promote binding of mRNA to ribosomes
by duplexing with the rRNA to allow correct positioning of the
ribosome. For a review on maximizing gene expression, see Roberts
and Lauer, Methods in Enzymology, 68:473 (1979), which is hereby
incorporated by reference in its entirety.
Promoters vary in their "strength" (i.e., their ability to promote
transcription). For the purposes of expressing a cloned gene, it is
desirable to use strong promoters in order to obtain a high level
of transcription and, hence, expression of the gene. Depending upon
the host cell system utilized, any one of a number of suitable
promoters may be used. For instance, when cloning in E. coli, its
bacteriophages, or plasmids, promoters such as the T7 phage
promoter, Zac promoter, trp promoter, recA promoter, ribosomal RNA
promoter, the P.sub.R and P.sub.L promoters of coliphage lambda and
others, including but not limited, to lacUV5, ompF, bla, lpp, and
the like, may be used to direct high levels of transcription of
adjacent DNA segments. Additionally, a hybrid trp-lacUV5 (tac)
promoter or other E. coli promoters produced by recombinant DNA or
other synthetic DNA techniques may be used to provide for
transcription of the inserted gene.
Bacterial host cell strains and expression vectors may be chosen
which inhibit the action of the promoter unless specifically
induced. In certain operons, the addition of specific inducers is
necessary for efficient transcription of the inserted DNA. For
example, the lac operon is induced by the addition of lactose or
IPTG (isopropylthio-beta-D-galactoside). A variety of other
operons, such as trp, pro, etc., are under different controls.
Specific initiation signals are also required for efficient gene
transcription and translation in prokaryotic cells. These
transcription and translation initiation signals may vary in
"strength" as measured by the quantity of gene specific messenger
RNA and protein synthesized, respectively. The DNA expression
vector, which contains a promoter, may also contain any combination
of various "strong" transcription and/or translation initiation
signals. For instance, efficient translation in E. coli requires a
Shine-Dalgarno ("SD") sequence about 7-9 bases 5' to the initiation
codon ("ATG") to provide a ribosome binding site. Thus, any SD-ATG
combination that can be utilized by host cell ribosomes may be
employed. Such combinations include, but are not limited to, the
SD-ATG combination from the cro gene or the N gene of coliphage
lambda, or from the E. coli tryptophan E, D, C, B or A genes.
Additionally, any SD-ATG combination produced by recombinant DNA or
other techniques involving incorporation of synthetic nucleotides
may be used.
Once the DNA molecule coding for a hypersensitive response elicitor
protein or polypeptide has been ligated to its appropriate
regulatory regions using well known molecular cloning techniques,
it can then be introduced into a vector or otherwise introduced
directly into a host cell (Sambrook et al., Molecular Cloning: A
Laboratory Manual, Second Edition, Cold Spring Harbor Press, N.Y.
(1989), which is hereby incorporated by reference in its entirety).
The recombinant molecule can be introduced into host cells via
transformation, particularly transduction, conjugation,
mobilization, or electroporation. Suitable host cells include, but
are not limited to, bacteria, virus, yeast, mammalian cells,
insect, plant, and the like. Preferably the host cells are either a
bacterial cell or a plant cell. The host cells, when grown in an
appropriate medium, are capable of expressing the hypersensitive
response elicitor protein or polypeptide, which can then be
isolated therefrom and, if necessary, purified.
Alternatively, it is desirable for recombinant host cells to
secrete the hypersensitive response elicitor protein or polypeptide
into growth medium, thereby avoiding the need to lyse cells and
remove cellular debris. To enable the host cell to secrete the
hypersensitive response elicitor, the host cell can also be
transformed with a type III secretion system in accordance with Ham
et al., "A Cloned Erwinia chrysanthemi Hrp (Type III Protein
Secretion) System Functions in Escherichia coli to Deliver
Pseudomonas syringae Avr Signals to Plant Cells and Secrete Avr
Proteins in Culture," Microbiol. 95:10206-10211 (1998), which is
hereby incorporated by reference in its entirety. After growing
recombinant host cells which secrete the hypersensitive response
elicitor into growth medium, isolation of the hypersensitive
response elicitor protein or polypeptide from growth medium can be
carried out substantially as described above.
The hypersensitive response elicitor of the present invention is
preferably in isolated form (i.e. separated from its host organism)
and more preferably produced in purified form (preferably at least
about 60%,) by conventional techniques. Typically, the
hypersensitive response elicitor of the present invention is
produced but not secreted into the growth medium of recombinant
host cells. Alternatively, the protein or polypeptide of the
present invention is secreted into growth medium. In the case of
unsecreted protein, to isolate the protein, the host cell (e.g., E.
coli) carrying a recombinant plasmid is propagated, lysed by
sonication, heat, or chemical treatment, and the homogenate is
centrifuged to remove bacterial debris. The supernatant is then
subjected to heat treatment and the hypersensitive response
elicitor is separated by centrifugation. The supernatant fraction
containing the hypersensitive response elicitor is subjected to gel
filtration in an appropriately sized dextran or polyacrylamide
column to separate the fragment. If necessary, the protein fraction
may be further purified by ion exchange or HPLC.
A composition suitable for treating plants or plant seeds with a
hypersensitive response elicitor polypeptide or protein in an
isolated form contains a hypersensitive response elicitor
polypeptide or protein in a carrier. Suitable carriers include
water, aqueous solutions, slurries, or dry powders. In this
embodiment, the composition contains greater than 500 nM
hypersensitive response elicitor polypeptide or protein.
Alternatively, application of the hypersensitive response elicitor
protein or polypeptide can also be applied in a non-isolated but
non-infectious form. When applied in non-isolated but
non-infectious form, the hypersensitive response elicitor is
applied indirectly to the plant via application of a bacteria which
expresses and then secretes or injects the expressed hypersensitive
response elicitor protein or polypeptide into plant cells or
tissues. Such application can be carried out by applying the
bacteria to all or part of a plant or a plant seed under conditions
where the polypeptide or protein contacts all or part of the cells
of the plant or plant seed. Alternatively, the hypersensitive
response elicitor protein or polypeptide can be applied to plants
such that seeds recovered from such plants themselves are able to
achieve the effects of the present invention.
In the bacterial application mode of the present invention, the
bacteria do not cause disease and have been transformed (e.g.,
recombinantly) with genes encoding a hypersensitive response
elicitor polypeptide or protein. For example, E. coli, which does
not elicit a hypersensitive response in plants, can be transformed
with genes encoding a hypersensitive response elicitor polypeptide
or protein and then applied to plants. Bacterial species other than
E. coli can also be used in this embodiment of the present
invention.
Alternatively, in the bacterial application mode of the present
invention, a naturally occurring virulent bacteria that is capable
of expressing and secreting a hypersensitive response elicitor is
mutated or altered to be an aviralent pathogen while retaining its
ability to express and secrete the hypersensitive response
elicitoris. Examples of such naturally occurring virulent bacteria
are noted above. In this embodiment, these bacteria are applied to
plants or their seeds. For example, virulent Erwinia amylovora
causes disease in apple. An avirulent Erwinia amylovora would not
cause the disease in apples, but would retain its ability to
express and secrete a hypersensitive response elicitor. Bacterial
species other than Erwinia amylovora can also be used in this
embodiment of the present invention.
The methods of the present invention which involve application of
the agricultural chemicals and/or hypersensitive response elicitor
polypeptides or proteins can be carried out through a variety of
procedures in which all or part of the plant is treated, including
leaves, stems, roots, etc. Application techniques may include but
not limited to; foliar application, broadcast application,
chemigation, high pressures injection, nesting, aerial spray,
utilization of chemstations, root drench, and cutting drench.
Application may, but need not, involve infiltration of the
hypersensitive response elicitor polypeptide or protein into the
plant. More than one application of the agricultural chemical
and/or hypersensitive response elicitor protein or polypeptide may
be desirable to realize maximal benefit over the course of a
growing season.
Agricultural chemicals and/or hypersensitive response elicitor
polypeptides or proteins can be applied to a plant or plant seed
alone or mixed with additional components. Additional components
can include one or more additional agricultural chemicals,
carriers, adjuvants, buffering agents, coating agents, abrading
agents, surfactants, preservatives, and color agents. These
materials can be used to facilitate the process of the present
invention. In addition, the agricultural chemicals and/or
hypersensitive response elicitor polypeptides or proteins can be
applied to plant seeds with other conventional seed formulation and
treatment materials, including clays and polysaccharides.
When treating plant seeds in accordance with the application
embodiment of the present invention, the agricultural chemicals
and/or hypersensitive response elicitor polypeptides or proteins
can be applied by low or high pressure spraying, seed dusting, seed
soaking, and seed coating, or injection. Other suitable application
procedures can be envisioned by those skilled in the art provided
they are able to effect contact of the hypersensitive response
elicitor polypeptide or protein with cells of the plant or plant
seed.
Once treated with the agricultural chemical and/or hypersensitive
response elicitor of the present invention, the seeds can be
planted in natural or artificial soil and cultivated using
conventional procedures to produce plants. After plants have been
propagated from seeds treated in accordance with the present
invention, the plants may also be treated with one or more
applications of the agricultural chemicals and/or hypersensitive
response elicitor polypeptides or proteins. Such propagated plants
may, in turn, be useful in producing seeds or propagules (e.g.,
cuttings) suitable for carrying out the present invention.
Typically, the manufacturer or distributor's product label for
specific agricultural chemicals and/or hypersensitive response
elicitor polypeptides or proteins will provide suggested
application rates, the crops on which use of the agricultural
chemicals and/or hypersensitive response elicitor polypeptides or
proteins has been approved, and preferred application techniques if
they exist.
The present method, for increasing the efficacy of common
agricultural chemicals, can be utilized while treating a wide
variety of plants and plant seeds types. Suitable plants include
dicots and monocots. More particularly, useful crop plants can
include, but are not limited to: canola, alfalfa, rice, wheat,
barley, rye, cotton, sunflower, peanut, corn, potato, sweet potato,
bean, pea, chicory, lettuce, endive, cabbage, brussel sprout, beet,
parsnip, cauliflower, broccoli, turnip, radish, spinach, onion,
garlic, eggplant, pepper, celery, carrot, squash, pumpkin,
zucchini, cucumber, apple, pear, melon, citrus, strawberry, grape,
raspberry, pineapple, soybean, tobacco, tomato, sorghum, and
sugarcane. Examples of suitable ornamental plants are: Arabidopsis
thaliana, Saintpaulia, petunia, pelargonium, poinsettia,
chrysanthemum, carnation, and zinnia.
In another embodiment of the present invention, one or more
agricultural chemicals are applied to a transgenic plants or
transgenic seeds encoding a hypersensitive response elicitor
protein or polypeptide. This technique involves the use of
transgenic plants and transgenic seeds encoding a hypersensitive
response elicitor protein or polypeptide, a hypersensitive response
elicitor proteins or polypeptides need not be applied to the plant
or seed. Instead, transgenic plants transformed with a gene
encoding such a hypersensitive response elicitor protein or
polypeptide are produced according to procedures well known in the
art as described below.
The vector described above can be microinjected directly into plant
cells by use of micropipettes to transfer mechanically the
recombinant DNA. Crossway, Mol. Gen. Genetics, 202:179-85 (1985),
which is hereby incorporated by reference in its entirety. The
genetic material may also be transferred into the plant cell using
polyethylene glycol. Krens, et al., Nature, 296:72-74 (1982), which
is hereby incorporated by reference in its entirety.
Another approach to transforming plant cells with a gene is
particle bombardment (also known as biolistic transformation) of
the host cell. This can be accomplished in one of several ways.
This technique is disclosed in U.S. Pat. Nos. 4,945,050, 5,036,006,
and 5,100,792, all to Sanford et al., which are hereby incorporated
by reference in their entirety. Generally, this procedure involves
propelling inert or biologically active particles at the cells
under conditions effective to penetrate the outer surface of the
cell and to be incorporated within the interior thereof When inert
particles are utilized, the vector can be introduced into the cell
by coating the particles with the vector containing the
heterologous DNA. Alternatively, the target cell can be surrounded
by the vector so that the vector is carried into the cell by the
wake of the particle. Biologically active particles (e.g., dried
bacterial cells containing the vector and heterologous DNA) can
also be propelled into plant cells.
Yet another method of introduction is fusion of protoplasts with
other entities, either minicells, cells, lysosomes, or other
fusible lipid-surfaced bodies. Fraley, et al., Proc. Natl. Acad.
Sci. USA, 79:1859-63 (1982), which is hereby incorporated by
reference in its entirety.
The DNA molecule may also be introduced into the plant cells by
electroporation. Fromm et al., Proc. Natl. Acad. Sci. USA, 82:5824
(1985), which is hereby incorporated by reference in its entirety.
In this technique, plant protoplasts are electroporated in the
presence of plasmids containing the expression cassette. Electrical
impulses of high field strength reversibly permeabilize
biomembranes allowing the introduction of the plasmids.
Electroporated plant protoplasts reform the cell wall, divide, and
regenerate.
Another method of introducing the DNA molecule into plant cells is
to infect a plant cell with Agrobacterium tumefaciens or A.
rhizogenes previously transformed with the gene. Under appropriate
conditions known in the art, the transformed plant cells are grown
to form shoots or roots, and develop further into plants.
Generally, this procedure involves inoculating the plant tissue
with a suspension of bacteria and incubating the tissue for 48 to
72 hours on regeneration medium without antibiotics at
25-28.degree. C.
Agrobacterium is a representative genus of the Gram-negative family
Rhizobiaceae. Its species are responsible for crown gall (A.
tumefaciens) and hairy root disease (A. rhizogenes). The plant
cells in crown gall tumors and hairy roots are induced to produce
amino acid derivatives known as opines, which are catabolized only
by the bacteria The bacterial genes responsible for expression of
opines are a convenient source of control elements for chimeric
expression cassettes. In addition, assaying for the presence of
opines can be used to identify transformed tissue.
Heterologous genetic sequences can be introduced into appropriate
plant cells, by means of the Ti plasmid of A. tumefaciens or the Ri
plasmid of A. rhizogenes. The Ti or Ri plasmid is transmitted to
plant cells on infection by Agrobacterium and is stably integrated
into the plant genome. J. Schell, Science, 237:1176-83 (1987),
which is hereby incorporated by reference in its entirety.
After transformation, the transformed plant cells must be
regenerated.
Plant regeneration from cultured protoplasts is described in Evans
et al., Handbook of Plant Cell Cultures, Vol. 1: (MacMillan
Publishing Co., New York, 1983); and Vasil I. R. (ed.), Cell
Culture and Somatic Cell Genetics of Plants, Acad. Press, Orlando,
Vol. I, 1984, and Vol. III (1986), which are hereby incorporated by
reference in their entirety.
It is known that practically all plants can be regenerated from
cultured cells or tissues, including but not limited to, all major
species of sugarcane, sugar beets, cotton, fruit trees, and
legumes.
Means for regeneration vary from species to species of plants, but
generally a suspension of transformed protoplasts or a petri plate
containing transformed explants is first provided. Callus tissue is
formed and shoots may be induced from callus and subsequently
rooted. Alternatively, embryo formation can be induced in the
callus tissue. These embryos germinate as natural embryos to form
plants. The culture media will generally contain various amino
acids and hormones, such as auxin and cytokinins. It is also
advantageous to add glutamic acid and proline to the medium,
especially for such species as corn and alfalfa. Efficient
regeneration will depend on the medium, on the genotype, and on the
history of the culture. If these three variables are controlled;
then regeneration is usually reproducible and repeatable.
After the expression cassette is stably incorporated in transgenic
plants, it can be transferred to other plants by sexual crossing.
Any of a number of standard breeding techniques can be used,
depending upon the species to be crossed.
Once transgenic plants of this type are produced, the plants
themselves can be cultivated in accordance with conventional
procedure. Alternatively, transgenic seeds or propagules (e.g.,
cuttings) are recovered from the transgenic plants. The seeds can
then be planted in the soil and cultivated using conventional
procedures to produce transgenic plants. The transgenic plants are
propagated from the planted transgenic seeds.
EXAMPLES
Example 1
Application of Messenger.RTM. with Roundup UltraMAX.RTM. to Improve
Control of Various Weeds
The objective of this study was to determine if pre, post, or
tank-mix application of Messenger (active ingredient harpin.sub.Ea)
affected Roundup UltraMAX's (active ingredient glyphosate,
Monsanto, St. Louis, Mo.) ability to control weeds. In this
experiment, control of two susceptible and two tolerant dicot weed
species, as well as two susceptible and two tolerant monocot weed
species was examined. Plots were constructed in the field and
uniformilly planted with the respective weed seeds. Plots were
maintained in ambient conditions. Messenger and Roundup UltraMAX
applications were conducted at 2.25 oz. per acre and 16 oz. per
acre, respectively. The various treatment groups were as follows;
(1) Messenger application followed three days later by a Roundup
UltraMAX application (Mess bf RU), (2) application of Messenger and
Roundup UltraMAX at the same time via a tank-mix (MSS+RU), (3)
application of Roundup UltraMAX followed one day (24 hours) later
by a Messenger application (RU bf MSS), (4) Roundup UltraMAX
application alone. Observations regarding the percent weed control
of the specific weed species were made at seven and 14 days after
treatments (DAT). Results are shown below in Tables 6 through
9.
TABLE-US-00020 TABLE 6 Effect of Messenger upon Roundup UltraMAX
Efficacy (susceptible dicots) Common Lambsquarter Common Cocklebur
Treatment 7 DAT 14 DAT 7 DAT 14 DAT MSS bf RU 62 b 82 b 82 b 100
MSS + RU 73 a 94 a 91 a 100 RU bf MSS 72 a 91 a 92 a 100 RU 45 c 72
c 72 c 100 Same letters do not significantly differ (P = .05,
Student-Newman-Keuls)
TABLE-US-00021 TABLE 7 Effect of Messenger upon Roundup UltraMAX
Efficacy (tolerant dicots) Velvetleaf Redroot Pigweed Treatment 7
DAT 14 DAT 7 DAT 14 DAT MSS bf RU 21 b 32 b 54 b 74 b MSS + RU 32 a
44 a 81 a 96 a RU bf MSS 33 a 46 a 77 a 94 a RU 11 c 18 c 35 c 46 c
Same letters do not significantly differ (P = .05,
Student-Newman-Keuls)
TABLE-US-00022 TABLE 8 Effect of Messenger upon Roundup UltraMAX
Efficacy (susceptible monocots) Smooth Crabgrass Giant Foxtail
Treatment 7 DAT 14 DAT 7 DAT 14 DAT MSS bf RU 80 b 100 83 b 100 MSS
+ RU 92 a 100 93 a 100 RU bf MSS 91 a 100 92 a 100 RU 72 c 100 75 c
100 Same letters do not significantly differ (P = .05,
Student-Newman-Keuls)
TABLE-US-00023 TABLE 9 Effect of Messenger upon Roundup UltraMAX
Efficacy (tolerant monocots) Yellow Nutsedge Shattercane Treatment
7 DAT 14 DAT 7 DAT 14 DAT MSS bf RU 5 b 10 c 42 b 70 b MSS + RU 14
a 29 a 75 a 97 a RU bf MSS 13 a 24 d 72 a 93 a RU 2 c 4 b 28 c 54 c
Same letters do not significantly differ (P = .05,
Student-Newman-Keuls)
In each case where 100% control was not achieved, the inclusion of
Messenger with Roundup UltraMAX significantly increased Roundup
UltraMAX's control of the weed. Though Messenger treatment followed
by Roundup UltraMAX treatment showed significantly increased weed
control over that of Roundup Ultra Max alone, tank-mixing and
application of Roundup UlItraMAX followed by Messenger application
showed the greatest control of weeds.
Example 2
Application of Messenger.RTM. with Orthene.RTM. to Control Insects
for Blue Mold in Tobacco Results in Lower Disease Incidence than
Orthene Alone
Tobacco (Nicotiana tobacum), var. K-326, was planted in a
small-plot, replicated (3 times) field trial. Application of
Messenger (active ingredient harpin.sub.Ea) Orthene (active
ingredient acephate, Valent U.S.A. Corp., Walnut Creek, Calif.),
and Messenger+Orthene were made beginning with the transplant water
and were followed by 4 foliar sprays at approximately 14-d
intervals. Orthene was used in this trial to control aphids, a
common vector for blue mold disease (Peronospora tabacina) in
tobacco.
The trial was not inoculated with insects or disease. Evaluation
for blue mold was made approximately one week following the final
(4.sup.th) foliar application of each treatment. Addition of
Messenger to the Orthene treatment resulted in lower blue mold
infestation than the Messenger alone treatment, while the
combination of both products resulted in substantially lower
disease incidence than the Orthene alone treatment (Table 10).
These results indicate a positive trend for the inclusion of
Messenger with Orthene to give a slightly greater disease control
than either Messenger or Orthene alone (Table 10).
TABLE-US-00024 TABLE 10 Messenger, Orthene, and Messenger + Orthene
treatments applied to tobacco as transplant water drenches (TPW)
and foliar sprays. APPL. RATE BLUE APPL. (FOLIAR MOLD DISEASE
TREATMENT(S) RATE (TPW) SPRAY) INCIDENCE (%) Messenger 30 ppm 30
ppm 8.2 Orthene 12 oz/A 12 oz/A 27.8 Messenger + 30 ppm + 30 ppm +
7.0 Orthene 12 oz/A 12 oz/A
Messenger vs. Messenger+Orthene: 15% decrease in blue mold disease
incidence.
Orthene vs. Messenger+Orthene: 75% decrease in blue mold disease
incidence.
Example 3
Application of Messenger.RTM. with Temik.RTM. to Control Nematodes
in Cotton Enhances Performance of Temik
Cotton, (Gossypium hirsutum), var. PM 1218, was planted to a
small-plot, replicated (6 times) field trial. Plot size was 6-8
rows.times.50 feet with the center 4 rows treated and center 2 rows
harvested. Ten-foot buffers were established between blocks. Temik
(active ingredient aldricarb, Aventis CropScience, Research
Triangle, N.C.) was applied in-furrow (at 5 lbs/A) at planting.
Messenger (active ingredient harpin.sub.Ea) foliar applications (at
2.23 oz/A) were made at various timing regimes on both
Temik-treated and non-Temik treated cotton. Yield data in response
to these treatments is shown in Table 11.
TABLE-US-00025 TABLE 11 Messenger, Temik, and Messenger + Temik
Treatments Effect on Cotton Seed Yield. SEED COTTOT SEED INCREASE
OVER TREATMENT YIELD (LBS/A) UNTREATED (%) Messenger 2,203.sup.1
8.9 Messenger + Temik 2,388.sup.1 18.0 Temik 2,221.sup. 9.8
Untreated 2,023.sup. -- .sup.1Seed cotton yield figures are
averages from four treatment-timing combinations of Messenger and
Messenger + Temik, respectively.
Results from this field trial indicated that both the individual
Messenger and Temik treatments boosted seed cotton yield about 10%
above the untreated. However, the Messenger+Temik treatment gave an
18% yield above the untreated suggesting that addition of Messenger
to the Temik treatment enhanced Temik's ability to perform its
intended function.
Example 4
Application of Messenger.RTM. with Equation Pro.RTM. to Control
Late Blight in Tomatoes Enhances Performance of Equation Pro
Tomato seedlings were planted into greenhouse pots, 3 plants per
pot replicated 4 pots per treatment. One week prior to artificial
inoculation with Phytopthora infestans (Late blight), one set of
plants received a single foliar spray of Messenger (active
ingredient harpin.sub.Ea) at approx. 20 ppm active ingredient
(a.i.) followed by a second foliar spray approximately one week
after inoculation. A second set of replicate pots received
Messenger+Equation Pro (active ingredients famoxadone+cymoxanil,
DuPont Crop Protection, Wilmington, Del.) while a third set of
replicates received only the Equation Pro treatment. An untreated
control treatment was included in the test. After the disease had
spread to fully infect the untreated plants, treated plants were
rated for disease symptoms; severity and index were both calculated
for each treatment. Results are presented in Table 12.
TABLE-US-00026 TABLE 12 Messenger, Messenger + Equation Pro, and
Equation Pro Treatments Effect on Late Blight in Tomato. DISEASE
SEVERITY EFFICACY TREATMENT INDEX (%) (%) Messenger 0.89.sup.1 17.9
71.0 Messenger + 0.30.sup.1 6.0 90.2 Equation Pro Equation Pro
0.59.sup. 11.8 80.8 Untreated 3.07.sup. 61.4 -- .sup.1Mean values
of four replicate pots, three plants in each.
Results from this greenhouse trial indicated that both the
individual Messenger and Equation Pro treatments provided
substantial resistance to Late blight in tomato. However, the
Messenger+Equation Pro treatment resulted in an even greater degree
of disease control than either treatment alone, suggesting that the
addition of Messenger to the Equation Pro treatment enhances
Equation Pro's ability to perform its intended function.
Example 5
Inclusion of Messenger.RTM. in Aliette.RTM. Treatment Program
Increases Control of Phytophthora cinnamomi Root Rot in Avocado
Five month old avocado seedlings (Topo Topa) were inoculated with
Phytophthora cinnamomi. Treatment groups included; (1) Aliette
(active ingredient fosetyl-aluminum ISO, Aventis CropScience,
Research Triangle Park, N.C.) pre-treatment, applied seven days
prior to inoculation, (2) Messenger (active ingredient
harpin.sub.Ea) treatments seven days prior to inoculation, 14 days
post-inoculation and every 21 days there after, (3) the combination
of treatments 1 and 2 described above, (4) inoculated untreated
control, and (5) uninoculated untreated control. Each treatment
group was replicated six times. Observations were recorded with
respect to the percent of necrotic roots present in the total root
mass. Avocado roots show a distinct blackening when infected with
P. cinnamomi, whereas non-infected roots are brown-white in color.
Table 13 summarizes the study details and resulting data.
TABLE-US-00027 TABLE 13 Messenger, Messenger + Aliette, and Aliette
Treatments Effect on Root Rot in Avocado. Treatment Application
Technique % Diseased Roots Aliette pre-treatment 60 bc Messenger
foliar every 21 days 38.3 c Aliette + Messenger pre-treat + foliar
21 d 27.5 cd UTC none 96.5 a UTC (no inoculation) none 6.3 d Same
letters do not significantly differ.
Although the invention has been described in detail for the purpose
of illustration, it is understood that such details are solely for
that purpose, and variations can be made therein by those skilled
in the art without departing from the spirit of the scope of the
invention which is defined by the following claims.
SEQUENCE LISTINGS
1
141338PRTErwinia chrysanthemi 1Met Gln Ile Thr Ile Lys Ala His Ile
Gly Gly Asp Leu Gly Val Ser1 5 10 15Gly Leu Gly Ala Gln Gly Leu Lys
Gly Leu Asn Ser Ala Ala Ser Ser 20 25 30Leu Gly Ser Ser Val Asp Lys
Leu Ser Ser Thr Ile Asp Lys Leu Thr 35 40 45Ser Ala Leu Thr Ser Met
Met Phe Gly Gly Ala Leu Ala Gln Gly Leu 50 55 60Gly Ala Ser Ser Lys
Gly Leu Gly Met Ser Asn Gln Leu Gly Gln Ser65 70 75 80Phe Gly Asn
Gly Ala Gln Gly Ala Ser Asn Leu Leu Ser Val Pro Lys 85 90 95Ser Gly
Gly Asp Ala Leu Ser Lys Met Phe Asp Lys Ala Leu Asp Asp 100 105
110Leu Leu Gly His Asp Thr Val Thr Lys Leu Thr Asn Gln Ser Asn Gln
115 120 125Leu Ala Asn Ser Met Leu Asn Ala Ser Gln Met Thr Gln Gly
Asn Met 130 135 140Asn Ala Phe Gly Ser Gly Val Asn Asn Ala Leu Ser
Ser Ile Leu Gly145 150 155 160Asn Gly Leu Gly Gln Ser Met Ser Gly
Phe Ser Gln Pro Ser Leu Gly 165 170 175Ala Gly Gly Leu Gln Gly Leu
Ser Gly Ala Gly Ala Phe Asn Gln Leu 180 185 190Gly Asn Ala Ile Gly
Met Gly Val Gly Gln Asn Ala Ala Leu Ser Ala 195 200 205Leu Ser Asn
Val Ser Thr His Val Asp Gly Asn Asn Arg His Phe Val 210 215 220Asp
Lys Glu Asp Arg Gly Met Ala Lys Glu Ile Gly Gln Phe Met Asp225 230
235 240Gln Tyr Pro Glu Ile Phe Gly Lys Pro Glu Tyr Gln Lys Asp Gly
Trp 245 250 255Ser Ser Pro Lys Thr Asp Asp Lys Ser Trp Ala Lys Ala
Leu Ser Lys 260 265 270Pro Asp Asp Asp Gly Met Thr Gly Ala Ser Met
Asp Lys Phe Arg Gln 275 280 285Ala Met Gly Met Ile Lys Ser Ala Val
Ala Gly Asp Thr Gly Asn Thr 290 295 300Asn Leu Asn Leu Arg Gly Ala
Gly Gly Ala Ser Leu Gly Ile Asp Ala305 310 315 320Ala Val Val Gly
Asp Lys Ile Ala Asn Met Ser Leu Gly Lys Leu Ala 325 330 335Asn
Ala22141DNAErwinia chrysanthemi 2cgattttacc cgggtgaacg tgctatgacc
gacagcatca cggtattcga caccgttacg 60gcgtttatgg ccgcgatgaa ccggcatcag
gcggcgcgct ggtcgccgca atccggcgtc 120gatctggtat ttcagtttgg
ggacaccggg cgtgaactca tgatgcagat tcagccgggg 180cagcaatatc
ccggcatgtt gcgcacgctg ctcgctcgtc gttatcagca ggcggcagag
240tgcgatggct gccatctgtg cctgaacggc agcgatgtat tgatcctctg
gtggccgctg 300ccgtcggatc ccggcagtta tccgcaggtg atcgaacgtt
tgtttgaact ggcgggaatg 360acgttgccgt cgctatccat agcaccgacg
gcgcgtccgc agacagggaa cggacgcgcc 420cgatcattaa gataaaggcg
gcttttttta ttgcaaaacg gtaacggtga ggaaccgttt 480caccgtcggc
gtcactcagt aacaagtatc catcatgatg cctacatcgg gatcggcgtg
540ggcatccgtt gcagatactt ttgcgaacac ctgacatgaa tgaggaaacg
aaattatgca 600aattacgatc aaagcgcaca tcggcggtga tttgggcgtc
tccggtctgg ggctgggtgc 660tcagggactg aaaggactga attccgcggc
ttcatcgctg ggttccagcg tggataaact 720gagcagcacc atcgataagt
tgacctccgc gctgacttcg atgatgtttg gcggcgcgct 780ggcgcagggg
ctgggcgcca gctcgaaggg gctggggatg agcaatcaac tgggccagtc
840tttcggcaat ggcgcgcagg gtgcgagcaa cctgctatcc gtaccgaaat
ccggcggcga 900tgcgttgtca aaaatgtttg ataaagcgct ggacgatctg
ctgggtcatg acaccgtgac 960caagctgact aaccagagca accaactggc
taattcaatg ctgaacgcca gccagatgac 1020ccagggtaat atgaatgcgt
tcggcagcgg tgtgaacaac gcactgtcgt ccattctcgg 1080caacggtctc
ggccagtcga tgagtggctt ctctcagcct tctctggggg caggcggctt
1140gcagggcctg agcggcgcgg gtgcattcaa ccagttgggt aatgccatcg
gcatgggcgt 1200ggggcagaat gctgcgctga gtgcgttgag taacgtcagc
acccacgtag acggtaacaa 1260ccgccacttt gtagataaag aagatcgcgg
catggcgaaa gagatcggcc agtttatgga 1320tcagtatccg gaaatattcg
gtaaaccgga ataccagaaa gatggctgga gttcgccgaa 1380gacggacgac
aaatcctggg ctaaagcgct gagtaaaccg gatgatgacg gtatgaccgg
1440cgccagcatg gacaaattcc gtcaggcgat gggtatgatc aaaagcgcgg
tggcgggtga 1500taccggcaat accaacctga acctgcgtgg cgcgggcggt
gcatcgctgg gtatcgatgc 1560ggctgtcgtc ggcgataaaa tagccaacat
gtcgctgggt aagctggcca acgcctgata 1620atctgtgctg gcctgataaa
gcggaaacga aaaaagagac ggggaagcct gtctcttttc 1680ttattatgcg
gtttatgcgg ttacctggac cggttaatca tcgtcatcga tctggtacaa
1740acgcacattt tcccgttcat tcgcgtcgtt acgcgccaca atcgcgatgg
catcttcctc 1800gtcgctcaga ttgcgcggct gatggggaac gccgggtgga
atatagagaa actcgccggc 1860cagatggaga cacgtctgcg ataaatctgt
gccgtaacgt gtttctatcc gcccctttag 1920cagatagatt gcggtttcgt
aatcaacatg gtaatgcggt tccgcctgtg cgccggccgg 1980gatcaccaca
atattcatag aaagctgtct tgcacctacc gtatcgcggg agataccgac
2040aaaatagggc agtttttgcg tggtatccgt ggggtgttcc ggcctgacaa
tcttgagttg 2100gttcgtcatc atctttctcc atctgggcga cctgatcggt t
21413403PRTErwinia amylovora - harpinEa 3Met Ser Leu Asn Thr Ser
Gly Leu Gly Ala Ser Thr Met Gln Ile Ser1 5 10 15Ile Gly Gly Ala Gly
Gly Asn Asn Gly Leu Leu Gly Thr Ser Arg Gln 20 25 30Asn Ala Gly Leu
Gly Gly Asn Ser Ala Leu Gly Leu Gly Gly Gly Asn 35 40 45Gln Asn Asp
Thr Val Asn Gln Leu Ala Gly Leu Leu Thr Gly Met Met 50 55 60Met Met
Met Ser Met Met Gly Gly Gly Gly Leu Met Gly Gly Gly Leu65 70 75
80Gly Gly Gly Leu Gly Asn Gly Leu Gly Gly Ser Gly Gly Leu Gly Glu
85 90 95Gly Leu Ser Asn Ala Leu Asn Asp Met Leu Gly Gly Ser Leu Asn
Thr 100 105 110Leu Gly Ser Lys Gly Gly Asn Asn Thr Thr Ser Thr Thr
Asn Ser Pro 115 120 125Leu Asp Gln Ala Leu Gly Ile Asn Ser Thr Ser
Gln Asn Asp Asp Ser 130 135 140Thr Ser Gly Thr Asp Ser Thr Ser Asp
Ser Ser Asp Pro Met Gln Gln145 150 155 160Leu Leu Lys Met Phe Ser
Glu Ile Met Gln Ser Leu Phe Gly Asp Gly 165 170 175Gln Asp Gly Thr
Gln Gly Ser Ser Ser Gly Gly Lys Gln Pro Thr Glu 180 185 190Gly Glu
Gln Asn Ala Tyr Lys Lys Gly Val Thr Asp Ala Leu Ser Gly 195 200
205Leu Met Gly Asn Gly Leu Ser Gln Leu Leu Gly Asn Gly Gly Leu Gly
210 215 220Gly Gly Gln Gly Gly Asn Ala Gly Thr Gly Leu Asp Gly Ser
Ser Leu225 230 235 240Gly Gly Lys Gly Leu Gln Asn Leu Ser Gly Pro
Val Asp Tyr Gln Gln 245 250 255Leu Gly Asn Ala Val Gly Thr Gly Ile
Gly Met Lys Ala Gly Ile Gln 260 265 270Ala Leu Asn Asp Ile Gly Thr
His Arg His Ser Ser Thr Arg Ser Phe 275 280 285Val Asn Lys Gly Asp
Arg Ala Met Ala Lys Glu Ile Gly Gln Phe Met 290 295 300Asp Gln Tyr
Pro Glu Val Phe Gly Lys Pro Gln Tyr Gln Lys Gly Pro305 310 315
320Gly Gln Glu Val Lys Thr Asp Asp Lys Ser Trp Ala Lys Ala Leu Ser
325 330 335Lys Pro Asp Asp Asp Gly Met Thr Pro Ala Ser Met Glu Gln
Phe Asn 340 345 350Lys Ala Lys Gly Met Ile Lys Arg Pro Met Ala Gly
Asp Thr Gly Asn 355 360 365Gly Asn Leu Gln Ala Arg Gly Ala Gly Gly
Ser Ser Leu Gly Ile Asp 370 375 380Ala Met Met Ala Gly Asp Ala Ile
Asn Asn Met Ala Leu Gly Lys Leu385 390 395 400Gly Ala
Ala41288DNAErwinia amylovora - harpinEa 4aagcttcggc atggcacgtt
tgaccgttgg gtcggcaggg tacgtttgaa ttattcataa 60gaggaatacg ttatgagtct
gaatacaagt gggctgggag cgtcaacgat gcaaatttct 120atcggcggtg
cgggcggaaa taacgggttg ctgggtacca gtcgccagaa tgctgggttg
180ggtggcaatt ctgcactggg gctgggcggc ggtaatcaaa atgataccgt
caatcagctg 240gctggcttac tcaccggcat gatgatgatg atgagcatga
tgggcggtgg tgggctgatg 300ggcggtggct taggcggtgg cttaggtaat
ggcttgggtg gctcaggtgg cctgggcgaa 360ggactgtcga acgcgctgaa
cgatatgtta ggcggttcgc tgaacacgct gggctcgaaa 420ggcggcaaca
ataccacttc aacaacaaat tccccgctgg accaggcgct gggtattaac
480tcaacgtccc aaaacgacga ttccacctcc ggcacagatt ccacctcaga
ctccagcgac 540ccgatgcagc agctgctgaa gatgttcagc gagataatgc
aaagcctgtt tggtgatggg 600caagatggca cccagggcag ttcctctggg
ggcaagcagc cgaccgaagg cgagcagaac 660gcctataaaa aaggagtcac
tgatgcgctg tcgggcctga tgggtaatgg tctgagccag 720ctccttggca
acgggggact gggaggtggt cagggcggta atgctggcac gggtcttgac
780ggttcgtcgc tgggcggcaa agggctgcaa aacctgagcg ggccggtgga
ctaccagcag 840ttaggtaacg ccgtgggtac cggtatcggt atgaaagcgg
gcattcaggc gctgaatgat 900atcggtacgc acaggcacag ttcaacccgt
tctttcgtca ataaaggcga tcgggcgatg 960gcgaaggaaa tcggtcagtt
catggaccag tatcctgagg tgtttggcaa gccgcagtac 1020cagaaaggcc
cgggtcagga ggtgaaaacc gatgacaaat catgggcaaa agcactgagc
1080aagccagatg acgacggaat gacaccagcc agtatggagc agttcaacaa
agccaagggc 1140atgatcaaaa ggcccatggc gggtgatacc ggcaacggca
acctgcaggc acgcggtgcc 1200ggtggttctt cgctgggtat tgatgccatg
atggccggtg atgccattaa caatatggca 1260cttggcaagc tgggcgcggc ttaagctt
12885447PRTErwinia amylovora 5Met Ser Ile Leu Thr Leu Asn Asn Asn
Thr Ser Ser Ser Pro Gly Leu1 5 10 15Phe Gln Ser Gly Gly Asp Asn Gly
Leu Gly Gly His Asn Ala Asn Ser 20 25 30Ala Leu Gly Gln Gln Pro Ile
Asp Arg Gln Thr Ile Glu Gln Met Ala 35 40 45Gln Leu Leu Ala Glu Leu
Leu Lys Ser Leu Leu Ser Pro Gln Ser Gly 50 55 60Asn Ala Ala Thr Gly
Ala Gly Gly Asn Asp Gln Thr Thr Gly Val Gly65 70 75 80Asn Ala Gly
Gly Leu Asn Gly Arg Lys Gly Thr Ala Gly Thr Thr Pro 85 90 95Gln Ser
Asp Ser Gln Asn Met Leu Ser Glu Met Gly Asn Asn Gly Leu 100 105
110Asp Gln Ala Ile Thr Pro Asp Gly Gln Gly Gly Gly Gln Ile Gly Asp
115 120 125Asn Pro Leu Leu Lys Ala Met Leu Lys Leu Ile Ala Arg Met
Met Asp 130 135 140Gly Gln Ser Asp Gln Phe Gly Gln Pro Gly Thr Gly
Asn Asn Ser Ala145 150 155 160Ser Ser Gly Thr Ser Ser Ser Gly Gly
Ser Pro Phe Asn Asp Leu Ser 165 170 175Gly Gly Lys Ala Pro Ser Gly
Asn Ser Pro Ser Gly Asn Tyr Ser Pro 180 185 190Val Ser Thr Phe Ser
Pro Pro Ser Thr Pro Thr Ser Pro Thr Ser Pro 195 200 205Leu Asp Phe
Pro Ser Ser Pro Thr Lys Ala Ala Gly Gly Ser Thr Pro 210 215 220Val
Thr Asp His Pro Asp Pro Val Gly Ser Ala Gly Ile Gly Ala Gly225 230
235 240Asn Ser Val Ala Phe Thr Ser Ala Gly Ala Asn Gln Thr Val Leu
His 245 250 255Asp Thr Ile Thr Val Lys Ala Gly Gln Val Phe Asp Gly
Lys Gly Gln 260 265 270Thr Phe Thr Ala Gly Ser Glu Leu Gly Asp Gly
Gly Gln Ser Glu Asn 275 280 285Gln Lys Pro Leu Phe Ile Leu Glu Asp
Gly Ala Ser Leu Lys Asn Val 290 295 300Thr Met Gly Asp Asp Gly Ala
Asp Gly Ile His Leu Tyr Gly Asp Ala305 310 315 320Lys Ile Asp Asn
Leu His Val Thr Asn Val Gly Glu Asp Ala Ile Thr 325 330 335Val Lys
Pro Asn Ser Ala Gly Lys Lys Ser His Val Glu Ile Thr Asn 340 345
350Ser Ser Phe Glu His Ala Ser Asp Lys Ile Leu Gln Leu Asn Ala Asp
355 360 365Thr Asn Leu Ser Val Asp Asn Val Lys Ala Lys Asp Phe Gly
Thr Phe 370 375 380Val Arg Thr Asn Gly Gly Gln Gln Gly Asn Trp Asp
Leu Asn Leu Ser385 390 395 400His Ile Ser Ala Glu Asp Gly Lys Phe
Ser Phe Val Lys Ser Asp Ser 405 410 415Glu Gly Leu Asn Val Asn Thr
Ser Asp Ile Ser Leu Gly Asp Val Glu 420 425 430Asn His Tyr Lys Val
Pro Met Ser Ala Asn Leu Lys Val Ala Glu 435 440 44561344DNAErwinia
amylovora 6atgtcaattc ttacgcttaa caacaatacc tcgtcctcgc cgggtctgtt
ccagtccggg 60ggggacaacg ggcttggtgg tcataatgca aattctgcgt tggggcaaca
acccatcgat 120cggcaaacca ttgagcaaat ggctcaatta ttggcggaac
tgttaaagtc actgctatcg 180ccacaatcag gtaatgcggc aaccggagcc
ggtggcaatg accagactac aggagttggt 240aacgctggcg gcctgaacgg
acgaaaaggc acagcaggaa ccactccgca gtctgacagt 300cagaacatgc
tgagtgagat gggcaacaac gggctggatc aggccatcac gcccgatggc
360cagggcggcg ggcagatcgg cgataatcct ttactgaaag ccatgctgaa
gcttattgca 420cgcatgatgg acggccaaag cgatcagttt ggccaacctg
gtacgggcaa caacagtgcc 480tcttccggta cttcttcatc tggcggttcc
ccttttaacg atctatcagg ggggaaggcc 540ccttccggca actccccttc
cggcaactac tctcccgtca gtaccttctc acccccatcc 600acgccaacgt
cccctacctc accgcttgat ttcccttctt ctcccaccaa agcagccggg
660ggcagcacgc cggtaaccga tcatcctgac cctgttggta gcgcgggcat
cggggccgga 720aattcggtgg ccttcaccag cgccggcgct aatcagacgg
tgctgcatga caccattacc 780gtgaaagcgg gtcaggtgtt tgatggcaaa
ggacaaacct tcaccgccgg ttcagaatta 840ggcgatggcg gccagtctga
aaaccagaaa ccgctgttta tactggaaga cggtgccagc 900ctgaaaaacg
tcaccatggg cgacgacggg gcggatggta ttcatcttta cggtgatgcc
960aaaatagaca atctgcacgt caccaacgtg ggtgaggacg cgattaccgt
taagccaaac 1020agcgcgggca aaaaatccca cgttgaaatc actaacagtt
ccttcgagca cgcctctgac 1080aagatcctgc agctgaatgc cgatactaac
ctgagcgttg acaacgtgaa ggccaaagac 1140tttggtactt ttgtacgcac
taacggcggt caacagggta actgggatct gaatctgagc 1200catatcagcg
cagaagacgg taagttctcg ttcgttaaaa gcgatagcga ggggctaaac
1260gtcaatacca gtgatatctc actgggtgat gttgaaaacc actacaaagt
gccgatgtcc 1320gccaacctga aggtggctga atga 13447341PRTPseudomonas
syringae 7Met Gln Ser Leu Ser Leu Asn Ser Ser Ser Leu Gln Thr Pro
Ala Met1 5 10 15Ala Leu Val Leu Val Arg Pro Glu Ala Glu Thr Thr Gly
Ser Thr Ser 20 25 30Ser Lys Ala Leu Gln Glu Val Val Val Lys Leu Ala
Glu Glu Leu Met 35 40 45Arg Asn Gly Gln Leu Asp Asp Ser Ser Pro Leu
Gly Lys Leu Leu Ala 50 55 60Lys Ser Met Ala Ala Asp Gly Lys Ala Gly
Gly Gly Ile Glu Asp Val65 70 75 80Ile Ala Ala Leu Asp Lys Leu Ile
His Glu Lys Leu Gly Asp Asn Phe 85 90 95Gly Ala Ser Ala Asp Ser Ala
Ser Gly Thr Gly Gln Gln Asp Leu Met 100 105 110Thr Gln Val Leu Asn
Gly Leu Ala Lys Ser Met Leu Asp Asp Leu Leu 115 120 125Thr Lys Gln
Asp Gly Gly Thr Ser Phe Ser Glu Asp Asp Met Pro Met 130 135 140Leu
Asn Lys Ile Ala Gln Phe Met Asp Asp Asn Pro Ala Gln Phe Pro145 150
155 160Lys Pro Asp Ser Gly Ser Trp Val Asn Glu Leu Lys Glu Asp Asn
Phe 165 170 175Leu Asp Gly Asp Glu Thr Ala Ala Phe Arg Ser Ala Leu
Asp Ile Ile 180 185 190Gly Gln Gln Leu Gly Asn Gln Gln Ser Asp Ala
Gly Ser Leu Ala Gly 195 200 205Thr Gly Gly Gly Leu Gly Thr Pro Ser
Ser Phe Ser Asn Asn Ser Ser 210 215 220Val Met Gly Asp Pro Leu Ile
Asp Ala Asn Thr Gly Pro Gly Asp Ser225 230 235 240Gly Asn Thr Arg
Gly Glu Ala Gly Gln Leu Ile Gly Glu Leu Ile Asp 245 250 255Arg Gly
Leu Gln Ser Val Leu Ala Gly Gly Gly Leu Gly Thr Pro Val 260 265
270Asn Thr Pro Gln Thr Gly Thr Ser Ala Asn Gly Gly Gln Ser Ala Gln
275 280 285Asp Leu Asp Gln Leu Leu Gly Gly Leu Leu Leu Lys Gly Leu
Glu Ala 290 295 300Thr Leu Lys Asp Ala Gly Gln Thr Gly Thr Asp Val
Gln Ser Ser Ala305 310 315 320Ala Gln Ile Ala Thr Leu Leu Val Ser
Thr Leu Leu Gln Gly Thr Arg 325 330 335Asn Gln Ala Ala Ala
34081026DNAPseudomonas syringae 8atgcagagtc tcagtcttaa cagcagctcg
ctgcaaaccc cggcaatggc ccttgtcctg 60gtacgtcctg aagccgagac gactggcagt
acgtcgagca aggcgcttca ggaagttgtc 120gtgaagctgg ccgaggaact
gatgcgcaat ggtcaactcg acgacagctc gccattggga 180aaactgttgg
ccaagtcgat ggccgcagat ggcaaggcgg gcggcggtat tgaggatgtc
240atcgctgcgc tggacaagct gatccatgaa aagctcggtg acaacttcgg
cgcgtctgcg 300gacagcgcct cgggtaccgg acagcaggac ctgatgactc
aggtgctcaa tggcctggcc 360aagtcgatgc tcgatgatct tctgaccaag
caggatggcg ggacaagctt ctccgaagac 420gatatgccga tgctgaacaa
gatcgcgcag ttcatggatg acaatcccgc acagtttccc 480aagccggact
cgggctcctg ggtgaacgaa ctcaaggaag acaacttcct tgatggcgac
540gaaacggctg cgttccgttc ggcactcgac atcattggcc agcaactggg
taatcagcag 600agtgacgctg gcagtctggc agggacgggt ggaggtctgg
gcactccgag cagtttttcc 660aacaactcgt ccgtgatggg
tgatccgctg atcgacgcca ataccggtcc cggtgacagc 720ggcaataccc
gtggtgaagc ggggcaactg atcggcgagc ttatcgaccg tggcctgcaa
780tcggtattgg ccggtggtgg actgggcaca cccgtaaaca ccccgcagac
cggtacgtcg 840gcgaatggcg gacagtccgc tcaggatctt gatcagttgc
tgggcggctt gctgctcaag 900ggcctggagg caacgctcaa ggatgccggg
caaacaggca ccgacgtgca gtcgagcgct 960gcgcaaatcg ccaccttgct
ggtcagtacg ctgctgcaag gcacccgcaa tcaggctgca 1020gcctga
10269424PRTPseudomonas syringae 9Met Ser Ile Gly Ile Thr Pro Arg
Pro Gln Gln Thr Thr Thr Pro Leu1 5 10 15Asp Phe Ser Ala Leu Ser Gly
Lys Ser Pro Gln Pro Asn Thr Phe Gly 20 25 30Glu Gln Asn Thr Gln Gln
Ala Ile Asp Pro Ser Ala Leu Leu Phe Gly 35 40 45Ser Asp Thr Gln Lys
Asp Val Asn Phe Gly Thr Pro Asp Ser Thr Val 50 55 60Gln Asn Pro Gln
Asp Ala Ser Lys Pro Asn Asp Ser Gln Ser Asn Ile65 70 75 80Ala Lys
Leu Ile Ser Ala Leu Ile Met Ser Leu Leu Gln Met Leu Thr 85 90 95Asn
Ser Asn Lys Lys Gln Asp Thr Asn Gln Glu Gln Pro Asp Ser Gln 100 105
110Ala Pro Phe Gln Asn Asn Gly Gly Leu Gly Thr Pro Ser Ala Asp Ser
115 120 125Gly Gly Gly Gly Thr Pro Asp Ala Thr Gly Gly Gly Gly Gly
Asp Thr 130 135 140Pro Ser Ala Thr Gly Gly Gly Gly Gly Asp Thr Pro
Thr Ala Thr Gly145 150 155 160Gly Gly Gly Ser Gly Gly Gly Gly Thr
Pro Thr Ala Thr Gly Gly Gly 165 170 175Ser Gly Gly Thr Pro Thr Ala
Thr Gly Gly Gly Glu Gly Gly Val Thr 180 185 190Pro Gln Ile Thr Pro
Gln Leu Ala Asn Pro Asn Arg Thr Ser Gly Thr 195 200 205Gly Ser Val
Ser Asp Thr Ala Gly Ser Thr Glu Gln Ala Gly Lys Ile 210 215 220Asn
Val Val Lys Asp Thr Ile Lys Val Gly Ala Gly Glu Val Phe Asp225 230
235 240Gly His Gly Ala Thr Phe Thr Ala Asp Lys Ser Met Gly Asn Gly
Asp 245 250 255Gln Gly Glu Asn Gln Lys Pro Met Phe Glu Leu Ala Glu
Gly Ala Thr 260 265 270Leu Lys Asn Val Asn Leu Gly Glu Asn Glu Val
Asp Gly Ile His Val 275 280 285Lys Ala Lys Asn Ala Gln Glu Val Thr
Ile Asp Asn Val His Ala Gln 290 295 300Asn Val Gly Glu Asp Leu Ile
Thr Val Lys Gly Glu Gly Gly Ala Ala305 310 315 320Val Thr Asn Leu
Asn Ile Lys Asn Ser Ser Ala Lys Gly Ala Asp Asp 325 330 335Lys Val
Val Gln Leu Asn Ala Asn Thr His Leu Lys Ile Asp Asn Phe 340 345
350Lys Ala Asp Asp Phe Gly Thr Met Val Arg Thr Asn Gly Gly Lys Gln
355 360 365Phe Asp Asp Met Ser Ile Glu Leu Asn Gly Ile Glu Ala Asn
His Gly 370 375 380Lys Phe Ala Leu Val Lys Ser Asp Ser Asp Asp Leu
Lys Leu Ala Thr385 390 395 400Gly Asn Ile Ala Met Thr Asp Val Lys
His Ala Tyr Asp Lys Thr Gln 405 410 415Ala Ser Thr Gln His Thr Glu
Leu 420101729DNAPseudomonas syringae 10tccacttcgc tgattttgaa
attggcagat tcatagaaac gttcaggtgt ggaaatcagg 60ctgagtgcgc agatttcgtt
gataagggtg tggtactggt cattgttggt catttcaagg 120cctctgagtg
cggtgcggag caataccagt cttcctgctg gcgtgtgcac actgagtcgc
180aggcataggc atttcagttc cttgcgttgg ttgggcatat aaaaaaagga
acttttaaaa 240acagtgcaat gagatgccgg caaaacggga accggtcgct
gcgctttgcc actcacttcg 300agcaagctca accccaaaca tccacatccc
tatcgaacgg acagcgatac ggccacttgc 360tctggtaaac cctggagctg
gcgtcggtcc aattgcccac ttagcgaggt aacgcagcat 420gagcatcggc
atcacacccc ggccgcaaca gaccaccacg ccactcgatt tttcggcgct
480aagcggcaag agtcctcaac caaacacgtt cggcgagcag aacactcagc
aagcgatcga 540cccgagtgca ctgttgttcg gcagcgacac acagaaagac
gtcaacttcg gcacgcccga 600cagcaccgtc cagaatccgc aggacgccag
caagcccaac gacagccagt ccaacatcgc 660taaattgatc agtgcattga
tcatgtcgtt gctgcagatg ctcaccaact ccaataaaaa 720gcaggacacc
aatcaggaac agcctgatag ccaggctcct ttccagaaca acggcgggct
780cggtacaccg tcggccgata gcgggggcgg cggtacaccg gatgcgacag
gtggcggcgg 840cggtgatacg ccaagcgcaa caggcggtgg cggcggtgat
actccgaccg caacaggcgg 900tggcggcagc ggtggcggcg gcacacccac
tgcaacaggt ggcggcagcg gtggcacacc 960cactgcaaca ggcggtggcg
agggtggcgt aacaccgcaa atcactccgc agttggccaa 1020ccctaaccgt
acctcaggta ctggctcggt gtcggacacc gcaggttcta ccgagcaagc
1080cggcaagatc aatgtggtga aagacaccat caaggtcggc gctggcgaag
tctttgacgg 1140ccacggcgca accttcactg ccgacaaatc tatgggtaac
ggagaccagg gcgaaaatca 1200gaagcccatg ttcgagctgg ctgaaggcgc
tacgttgaag aatgtgaacc tgggtgagaa 1260cgaggtcgat ggcatccacg
tgaaagccaa aaacgctcag gaagtcacca ttgacaacgt 1320gcatgcccag
aacgtcggtg aagacctgat tacggtcaaa ggcgagggag gcgcagcggt
1380cactaatctg aacatcaaga acagcagtgc caaaggtgca gacgacaagg
ttgtccagct 1440caacgccaac actcacttga aaatcgacaa cttcaaggcc
gacgatttcg gcacgatggt 1500tcgcaccaac ggtggcaagc agtttgatga
catgagcatc gagctgaacg gcatcgaagc 1560taaccacggc aagttcgccc
tggtgaaaag cgacagtgac gatctgaagc tggcaacggg 1620caacatcgcc
atgaccgacg tcaaacacgc ctacgataaa acccaggcat cgacccaaca
1680caccgagctt tgaatccaga caagtagctt gaaaaaaggg ggtggactc
172911344PRTPseudomonas solanacearum 11Met Ser Val Gly Asn Ile Gln
Ser Pro Ser Asn Leu Pro Gly Leu Gln1 5 10 15Asn Leu Asn Leu Asn Thr
Asn Thr Asn Ser Gln Gln Ser Gly Gln Ser 20 25 30Val Gln Asp Leu Ile
Lys Gln Val Glu Lys Asp Ile Leu Asn Ile Ile 35 40 45Ala Ala Leu Val
Gln Lys Ala Ala Gln Ser Ala Gly Gly Asn Thr Gly 50 55 60Asn Thr Gly
Asn Ala Pro Ala Lys Asp Gly Asn Ala Asn Ala Gly Ala65 70 75 80Asn
Asp Pro Ser Lys Asn Asp Pro Ser Lys Ser Gln Ala Pro Gln Ser 85 90
95Ala Asn Lys Thr Gly Asn Val Asp Asp Ala Asn Asn Gln Asp Pro Met
100 105 110Gln Ala Leu Met Gln Leu Leu Glu Asp Leu Val Lys Leu Leu
Lys Ala 115 120 125Ala Leu His Met Gln Gln Pro Gly Gly Asn Asp Lys
Gly Asn Gly Val 130 135 140Gly Gly Ala Asn Gly Ala Lys Gly Ala Gly
Gly Gln Gly Gly Leu Ala145 150 155 160Glu Ala Leu Gln Glu Ile Glu
Gln Ile Leu Ala Gln Leu Gly Gly Gly 165 170 175Gly Ala Gly Ala Gly
Gly Ala Gly Gly Gly Val Gly Gly Ala Gly Gly 180 185 190Ala Asp Gly
Gly Ser Gly Ala Gly Gly Ala Gly Gly Ala Asn Gly Ala 195 200 205Asp
Gly Gly Asn Gly Val Asn Gly Asn Gln Ala Asn Gly Pro Gln Asn 210 215
220Ala Gly Asp Val Asn Gly Ala Asn Gly Ala Asp Asp Gly Ser Glu
Asp225 230 235 240Gln Gly Gly Leu Thr Gly Val Leu Gln Lys Leu Met
Lys Ile Leu Asn 245 250 255Ala Leu Val Gln Met Met Gln Gln Gly Gly
Leu Gly Gly Gly Asn Gln 260 265 270Ala Gln Gly Gly Ser Lys Gly Ala
Gly Asn Ala Ser Pro Ala Ser Gly 275 280 285Ala Asn Pro Gly Ala Asn
Gln Pro Gly Ser Ala Asp Asp Gln Ser Ser 290 295 300Gly Gln Asn Asn
Leu Gln Ser Gln Ile Met Asp Val Val Lys Glu Val305 310 315 320Val
Gln Ile Leu Gln Gln Met Leu Ala Ala Gln Asn Gly Gly Ser Gln 325 330
335Gln Ser Thr Ser Thr Gln Pro Met 340121035DNAPseudomonas
solanacearum 12atgtcagtcg gaaacatcca gagcccgtcg aacctcccgg
gtctgcagaa cctgaacctc 60aacaccaaca ccaacagcca gcaatcgggc cagtccgtgc
aagacctgat caagcaggtc 120gagaaggaca tcctcaacat catcgcagcc
ctcgtgcaga aggccgcaca gtcggcgggc 180ggcaacaccg gtaacaccgg
caacgcgccg gcgaaggacg gcaatgccaa cgcgggcgcc 240aacgacccga
gcaagaacga cccgagcaag agccaggctc cgcagtcggc caacaagacc
300ggcaacgtcg acgacgccaa caaccaggat ccgatgcaag cgctgatgca
gctgctggaa 360gacctggtga agctgctgaa ggcggccctg cacatgcagc
agcccggcgg caatgacaag 420ggcaacggcg tgggcggtgc caacggcgcc
aagggtgccg gcggccaggg cggcctggcc 480gaagcgctgc aggagatcga
gcagatcctc gcccagctcg gcggcggcgg tgctggcgcc 540ggcggcgcgg
gtggcggtgt cggcggtgct ggtggcgcgg atggcggctc cggtgcgggt
600ggcgcaggcg gtgcgaacgg cgccgacggc ggcaatggcg tgaacggcaa
ccaggcgaac 660ggcccgcaga acgcaggcga tgtcaacggt gccaacggcg
cggatgacgg cagcgaagac 720cagggcggcc tcaccggcgt gctgcaaaag
ctgatgaaga tcctgaacgc gctggtgcag 780atgatgcagc aaggcggcct
cggcggcggc aaccaggcgc agggcggctc gaagggtgcc 840ggcaacgcct
cgccggcttc cggcgcgaac ccgggcgcga accagcccgg ttcggcggat
900gatcaatcgt ccggccagaa caatctgcaa tcccagatca tggatgtggt
gaaggaggtc 960gtccagatcc tgcagcagat gctggcggcg cagaacggcg
gcagccagca gtccacctcg 1020acgcagccga tgtaa 103513114PRTXanthomonas
campestris 13Met Asp Ser Ile Gly Asn Asn Phe Ser Asn Ile Gly Asn
Leu Gln Thr1 5 10 15Met Gly Ile Gly Pro Gln Gln His Glu Asp Ser Ser
Gln Gln Ser Pro 20 25 30Ser Ala Gly Ser Glu Gln Gln Leu Asp Gln Leu
Leu Ala Met Phe Ile 35 40 45Met Met Met Leu Gln Gln Ser Gln Gly Ser
Asp Ala Asn Gln Glu Cys 50 55 60Gly Asn Glu Gln Pro Gln Asn Gly Gln
Gln Glu Gly Leu Ser Pro Leu65 70 75 80Thr Gln Met Leu Met Gln Ile
Val Met Gln Leu Met Gln Asn Gln Gly 85 90 95Gly Ala Gly Met Gly Gly
Gly Gly Ser Val Asn Ser Ser Leu Gly Gly 100 105 110Asn Ala
14342DNAXanthomonas campestris 14atggactcta tcggaaacaa cttttcgaat
atcggcaacc tgcagacgat gggcatcggg 60cctcagcaac acgaggactc cagccagcag
tcgccttcgg ctggctccga gcagcagctg 120gatcagttgc tcgccatgtt
catcatgatg atgctgcaac agagccaggg cagcgatgca 180aatcaggagt
gtggcaacga acaaccgcag aacggtcaac aggaaggcct gagtccgttg
240acgcagatgc tgatgcagat cgtgatgcag ctgatgcaga accagggcgg
cgccggcatg 300ggcggtggcg gttcggtcaa cagcagcctg ggcggcaacg cc
342
* * * * *